diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,5188 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9976711690731253, + "eval_steps": 500, + "global_step": 3219, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.004657661853749418, + "grad_norm": 0.3246580015338979, + "learning_rate": 7.763975155279503e-07, + "loss": 0.4054, + "num_tokens": 2621440.0, + "step": 5 + }, + { + "epoch": 0.009315323707498836, + "grad_norm": 0.30502580431160947, + "learning_rate": 1.5527950310559006e-06, + "loss": 0.4067, + "num_tokens": 5242880.0, + "step": 10 + }, + { + "epoch": 0.013972985561248253, + "grad_norm": 0.24294102073505805, + "learning_rate": 2.329192546583851e-06, + "loss": 0.3997, + "num_tokens": 7845719.0, + "step": 15 + }, + { + "epoch": 0.018630647414997672, + "grad_norm": 0.29669820095143357, + "learning_rate": 3.1055900621118013e-06, + "loss": 0.4053, + "num_tokens": 10457626.0, + "step": 20 + }, + { + "epoch": 0.02328830926874709, + "grad_norm": 0.27363318055880104, + "learning_rate": 3.881987577639752e-06, + "loss": 0.4005, + "num_tokens": 13067414.0, + "step": 25 + }, + { + "epoch": 0.027945971122496506, + "grad_norm": 0.2906913628187444, + "learning_rate": 4.658385093167702e-06, + "loss": 0.4053, + "num_tokens": 15688854.0, + "step": 30 + }, + { + "epoch": 0.032603632976245925, + "grad_norm": 0.26831415568112743, + "learning_rate": 5.4347826086956525e-06, + "loss": 0.4097, + "num_tokens": 18278844.0, + "step": 35 + }, + { + "epoch": 0.037261294829995344, + "grad_norm": 0.24707490474424423, + "learning_rate": 6.2111801242236025e-06, + "loss": 0.4125, + "num_tokens": 20900161.0, + "step": 40 + }, + { + "epoch": 0.04191895668374476, + "grad_norm": 0.25285098310366294, + "learning_rate": 6.9875776397515525e-06, + "loss": 0.3876, + "num_tokens": 23521601.0, + "step": 45 + }, + { + "epoch": 0.04657661853749418, + "grad_norm": 0.25343162367275623, + "learning_rate": 7.763975155279503e-06, + "loss": 0.3943, + "num_tokens": 26143041.0, + "step": 50 + }, + { + "epoch": 0.05123428039124359, + "grad_norm": 0.23791425487205492, + "learning_rate": 8.540372670807453e-06, + "loss": 0.392, + "num_tokens": 28764481.0, + "step": 55 + }, + { + "epoch": 0.05589194224499301, + "grad_norm": 0.22651488429822728, + "learning_rate": 9.316770186335403e-06, + "loss": 0.3917, + "num_tokens": 31385921.0, + "step": 60 + }, + { + "epoch": 0.06054960409874243, + "grad_norm": 0.25448337996378584, + "learning_rate": 1.0093167701863353e-05, + "loss": 0.3882, + "num_tokens": 33990177.0, + "step": 65 + }, + { + "epoch": 0.06520726595249185, + "grad_norm": 0.22131771878749865, + "learning_rate": 1.0869565217391305e-05, + "loss": 0.3744, + "num_tokens": 36611617.0, + "step": 70 + }, + { + "epoch": 0.06986492780624126, + "grad_norm": 0.2669486917080392, + "learning_rate": 1.1645962732919255e-05, + "loss": 0.3782, + "num_tokens": 39218907.0, + "step": 75 + }, + { + "epoch": 0.07452258965999069, + "grad_norm": 0.25872212177529486, + "learning_rate": 1.2422360248447205e-05, + "loss": 0.3903, + "num_tokens": 41840347.0, + "step": 80 + }, + { + "epoch": 0.0791802515137401, + "grad_norm": 0.22749950020081922, + "learning_rate": 1.3198757763975155e-05, + "loss": 0.3722, + "num_tokens": 44461787.0, + "step": 85 + }, + { + "epoch": 0.08383791336748952, + "grad_norm": 0.2850975365464826, + "learning_rate": 1.3975155279503105e-05, + "loss": 0.3785, + "num_tokens": 47083227.0, + "step": 90 + }, + { + "epoch": 0.08849557522123894, + "grad_norm": 0.24567493862103307, + "learning_rate": 1.4751552795031057e-05, + "loss": 0.3817, + "num_tokens": 49704667.0, + "step": 95 + }, + { + "epoch": 0.09315323707498836, + "grad_norm": 0.25985197139962063, + "learning_rate": 1.5527950310559007e-05, + "loss": 0.3765, + "num_tokens": 52326107.0, + "step": 100 + }, + { + "epoch": 0.09781089892873777, + "grad_norm": 0.2910146307916693, + "learning_rate": 1.630434782608696e-05, + "loss": 0.3756, + "num_tokens": 54947547.0, + "step": 105 + }, + { + "epoch": 0.10246856078248719, + "grad_norm": 0.2775046213474384, + "learning_rate": 1.7080745341614907e-05, + "loss": 0.3697, + "num_tokens": 57568987.0, + "step": 110 + }, + { + "epoch": 0.10712622263623661, + "grad_norm": 0.348005475178152, + "learning_rate": 1.785714285714286e-05, + "loss": 0.3687, + "num_tokens": 60190427.0, + "step": 115 + }, + { + "epoch": 0.11178388448998602, + "grad_norm": 0.29715452279773225, + "learning_rate": 1.8633540372670807e-05, + "loss": 0.3786, + "num_tokens": 62811867.0, + "step": 120 + }, + { + "epoch": 0.11644154634373545, + "grad_norm": 0.30874357889032844, + "learning_rate": 1.940993788819876e-05, + "loss": 0.3612, + "num_tokens": 65433307.0, + "step": 125 + }, + { + "epoch": 0.12109920819748486, + "grad_norm": 0.31748202342003423, + "learning_rate": 2.0186335403726707e-05, + "loss": 0.3611, + "num_tokens": 68034228.0, + "step": 130 + }, + { + "epoch": 0.1257568700512343, + "grad_norm": 0.33434897463305013, + "learning_rate": 2.096273291925466e-05, + "loss": 0.3683, + "num_tokens": 70655668.0, + "step": 135 + }, + { + "epoch": 0.1304145319049837, + "grad_norm": 0.30833968968841147, + "learning_rate": 2.173913043478261e-05, + "loss": 0.3613, + "num_tokens": 73277108.0, + "step": 140 + }, + { + "epoch": 0.1350721937587331, + "grad_norm": 0.3231346578790754, + "learning_rate": 2.2515527950310562e-05, + "loss": 0.3693, + "num_tokens": 75898548.0, + "step": 145 + }, + { + "epoch": 0.13972985561248252, + "grad_norm": 0.3529022033943459, + "learning_rate": 2.329192546583851e-05, + "loss": 0.3626, + "num_tokens": 78519988.0, + "step": 150 + }, + { + "epoch": 0.14438751746623196, + "grad_norm": 0.3371933430263348, + "learning_rate": 2.4068322981366462e-05, + "loss": 0.3495, + "num_tokens": 81141428.0, + "step": 155 + }, + { + "epoch": 0.14904517931998137, + "grad_norm": 0.34848324489455, + "learning_rate": 2.484472049689441e-05, + "loss": 0.3531, + "num_tokens": 83733147.0, + "step": 160 + }, + { + "epoch": 0.1537028411737308, + "grad_norm": 0.383127789784734, + "learning_rate": 2.5621118012422362e-05, + "loss": 0.3525, + "num_tokens": 86351565.0, + "step": 165 + }, + { + "epoch": 0.1583605030274802, + "grad_norm": 0.348878572043702, + "learning_rate": 2.639751552795031e-05, + "loss": 0.3512, + "num_tokens": 88973005.0, + "step": 170 + }, + { + "epoch": 0.1630181648812296, + "grad_norm": 0.40565527852072275, + "learning_rate": 2.7173913043478262e-05, + "loss": 0.3595, + "num_tokens": 91572965.0, + "step": 175 + }, + { + "epoch": 0.16767582673497905, + "grad_norm": 0.29809752183694505, + "learning_rate": 2.795031055900621e-05, + "loss": 0.3521, + "num_tokens": 94194405.0, + "step": 180 + }, + { + "epoch": 0.17233348858872846, + "grad_norm": 0.46897573673807097, + "learning_rate": 2.8726708074534165e-05, + "loss": 0.3522, + "num_tokens": 96814074.0, + "step": 185 + }, + { + "epoch": 0.17699115044247787, + "grad_norm": 0.3700763516693258, + "learning_rate": 2.9503105590062114e-05, + "loss": 0.3436, + "num_tokens": 99435514.0, + "step": 190 + }, + { + "epoch": 0.18164881229622729, + "grad_norm": 0.36250511200009977, + "learning_rate": 3.0279503105590062e-05, + "loss": 0.3487, + "num_tokens": 102056954.0, + "step": 195 + }, + { + "epoch": 0.18630647414997673, + "grad_norm": 0.4535239213643802, + "learning_rate": 3.1055900621118014e-05, + "loss": 0.326, + "num_tokens": 104633209.0, + "step": 200 + }, + { + "epoch": 0.19096413600372614, + "grad_norm": 0.4326434738607519, + "learning_rate": 3.183229813664597e-05, + "loss": 0.3402, + "num_tokens": 107177869.0, + "step": 205 + }, + { + "epoch": 0.19562179785747555, + "grad_norm": 0.4383076498405631, + "learning_rate": 3.260869565217392e-05, + "loss": 0.3316, + "num_tokens": 109799309.0, + "step": 210 + }, + { + "epoch": 0.20027945971122496, + "grad_norm": 0.5145313193864417, + "learning_rate": 3.3385093167701865e-05, + "loss": 0.338, + "num_tokens": 112389962.0, + "step": 215 + }, + { + "epoch": 0.20493712156497437, + "grad_norm": 0.3674584315687584, + "learning_rate": 3.4161490683229814e-05, + "loss": 0.336, + "num_tokens": 115000610.0, + "step": 220 + }, + { + "epoch": 0.2095947834187238, + "grad_norm": 0.46762626808542057, + "learning_rate": 3.493788819875777e-05, + "loss": 0.3363, + "num_tokens": 117622050.0, + "step": 225 + }, + { + "epoch": 0.21425244527247322, + "grad_norm": 0.378012859125403, + "learning_rate": 3.571428571428572e-05, + "loss": 0.3333, + "num_tokens": 120229718.0, + "step": 230 + }, + { + "epoch": 0.21891010712622264, + "grad_norm": 0.4281393022990085, + "learning_rate": 3.6490683229813665e-05, + "loss": 0.3266, + "num_tokens": 122851158.0, + "step": 235 + }, + { + "epoch": 0.22356776897997205, + "grad_norm": 0.570048871418008, + "learning_rate": 3.7267080745341614e-05, + "loss": 0.3287, + "num_tokens": 125472598.0, + "step": 240 + }, + { + "epoch": 0.22822543083372146, + "grad_norm": 0.5343887029456514, + "learning_rate": 3.804347826086957e-05, + "loss": 0.3156, + "num_tokens": 128065220.0, + "step": 245 + }, + { + "epoch": 0.2328830926874709, + "grad_norm": 0.4625893787751897, + "learning_rate": 3.881987577639752e-05, + "loss": 0.3156, + "num_tokens": 130686660.0, + "step": 250 + }, + { + "epoch": 0.2375407545412203, + "grad_norm": 0.43218634350538265, + "learning_rate": 3.9596273291925465e-05, + "loss": 0.3204, + "num_tokens": 133297528.0, + "step": 255 + }, + { + "epoch": 0.24219841639496972, + "grad_norm": 0.4186876218844114, + "learning_rate": 4.0372670807453414e-05, + "loss": 0.3219, + "num_tokens": 135918968.0, + "step": 260 + }, + { + "epoch": 0.24685607824871914, + "grad_norm": 0.5702442122844179, + "learning_rate": 4.114906832298137e-05, + "loss": 0.3184, + "num_tokens": 138540408.0, + "step": 265 + }, + { + "epoch": 0.2515137401024686, + "grad_norm": 0.4360821769531783, + "learning_rate": 4.192546583850932e-05, + "loss": 0.3091, + "num_tokens": 141161848.0, + "step": 270 + }, + { + "epoch": 0.25617140195621796, + "grad_norm": 0.4520415095190347, + "learning_rate": 4.270186335403727e-05, + "loss": 0.3189, + "num_tokens": 143774319.0, + "step": 275 + }, + { + "epoch": 0.2608290638099674, + "grad_norm": 0.4867584283827858, + "learning_rate": 4.347826086956522e-05, + "loss": 0.3072, + "num_tokens": 146395759.0, + "step": 280 + }, + { + "epoch": 0.26548672566371684, + "grad_norm": 0.531950209741864, + "learning_rate": 4.425465838509317e-05, + "loss": 0.3098, + "num_tokens": 149017199.0, + "step": 285 + }, + { + "epoch": 0.2701443875174662, + "grad_norm": 0.4196291051314364, + "learning_rate": 4.5031055900621124e-05, + "loss": 0.3052, + "num_tokens": 151638639.0, + "step": 290 + }, + { + "epoch": 0.27480204937121566, + "grad_norm": 0.6527165477900265, + "learning_rate": 4.580745341614907e-05, + "loss": 0.3016, + "num_tokens": 154260079.0, + "step": 295 + }, + { + "epoch": 0.27945971122496505, + "grad_norm": 0.5379025127643137, + "learning_rate": 4.658385093167702e-05, + "loss": 0.3068, + "num_tokens": 156881519.0, + "step": 300 + }, + { + "epoch": 0.2841173730787145, + "grad_norm": 0.6496049826211475, + "learning_rate": 4.736024844720497e-05, + "loss": 0.3088, + "num_tokens": 159470618.0, + "step": 305 + }, + { + "epoch": 0.2887750349324639, + "grad_norm": 0.46608959595914534, + "learning_rate": 4.8136645962732924e-05, + "loss": 0.3043, + "num_tokens": 162092058.0, + "step": 310 + }, + { + "epoch": 0.2934326967862133, + "grad_norm": 0.4688858785603408, + "learning_rate": 4.891304347826087e-05, + "loss": 0.3025, + "num_tokens": 164713498.0, + "step": 315 + }, + { + "epoch": 0.29809035863996275, + "grad_norm": 0.5272676153698671, + "learning_rate": 4.968944099378882e-05, + "loss": 0.3059, + "num_tokens": 167334938.0, + "step": 320 + }, + { + "epoch": 0.30274802049371213, + "grad_norm": 0.630120084552457, + "learning_rate": 4.994822229892993e-05, + "loss": 0.3079, + "num_tokens": 169956378.0, + "step": 325 + }, + { + "epoch": 0.3074056823474616, + "grad_norm": 0.8807680093317907, + "learning_rate": 4.986192613047981e-05, + "loss": 0.3053, + "num_tokens": 172577818.0, + "step": 330 + }, + { + "epoch": 0.312063344201211, + "grad_norm": 0.4762589007693949, + "learning_rate": 4.977562996202969e-05, + "loss": 0.3024, + "num_tokens": 175199258.0, + "step": 335 + }, + { + "epoch": 0.3167210060549604, + "grad_norm": 1.4269007218795318, + "learning_rate": 4.968933379357957e-05, + "loss": 0.3232, + "num_tokens": 177820698.0, + "step": 340 + }, + { + "epoch": 0.32137866790870984, + "grad_norm": 0.45352771708953626, + "learning_rate": 4.9603037625129445e-05, + "loss": 0.3053, + "num_tokens": 180442138.0, + "step": 345 + }, + { + "epoch": 0.3260363297624592, + "grad_norm": 0.5285022339179731, + "learning_rate": 4.951674145667933e-05, + "loss": 0.3036, + "num_tokens": 183063578.0, + "step": 350 + }, + { + "epoch": 0.33069399161620866, + "grad_norm": 0.5169751746621581, + "learning_rate": 4.94304452882292e-05, + "loss": 0.2974, + "num_tokens": 185685018.0, + "step": 355 + }, + { + "epoch": 0.3353516534699581, + "grad_norm": 0.548293496106218, + "learning_rate": 4.934414911977908e-05, + "loss": 0.3039, + "num_tokens": 188300024.0, + "step": 360 + }, + { + "epoch": 0.3400093153237075, + "grad_norm": 0.4295101153028381, + "learning_rate": 4.9257852951328965e-05, + "loss": 0.2986, + "num_tokens": 190921464.0, + "step": 365 + }, + { + "epoch": 0.3446669771774569, + "grad_norm": 0.417345362400268, + "learning_rate": 4.917155678287884e-05, + "loss": 0.3092, + "num_tokens": 193536988.0, + "step": 370 + }, + { + "epoch": 0.3493246390312063, + "grad_norm": 0.5131381678527972, + "learning_rate": 4.908526061442872e-05, + "loss": 0.3034, + "num_tokens": 196158428.0, + "step": 375 + }, + { + "epoch": 0.35398230088495575, + "grad_norm": 0.5052096528443951, + "learning_rate": 4.89989644459786e-05, + "loss": 0.2947, + "num_tokens": 198779868.0, + "step": 380 + }, + { + "epoch": 0.3586399627387052, + "grad_norm": 0.4356730038516669, + "learning_rate": 4.891266827752848e-05, + "loss": 0.2912, + "num_tokens": 201401308.0, + "step": 385 + }, + { + "epoch": 0.36329762459245457, + "grad_norm": 0.5253395031021396, + "learning_rate": 4.882637210907836e-05, + "loss": 0.2944, + "num_tokens": 204007207.0, + "step": 390 + }, + { + "epoch": 0.367955286446204, + "grad_norm": 0.5029033782379033, + "learning_rate": 4.874007594062824e-05, + "loss": 0.291, + "num_tokens": 206628647.0, + "step": 395 + }, + { + "epoch": 0.37261294829995345, + "grad_norm": 0.4270568679976557, + "learning_rate": 4.865377977217811e-05, + "loss": 0.295, + "num_tokens": 209250087.0, + "step": 400 + }, + { + "epoch": 0.37727061015370283, + "grad_norm": 0.3464954396580666, + "learning_rate": 4.8567483603728e-05, + "loss": 0.2928, + "num_tokens": 211846391.0, + "step": 405 + }, + { + "epoch": 0.3819282720074523, + "grad_norm": 0.4624178520334951, + "learning_rate": 4.8481187435277875e-05, + "loss": 0.2931, + "num_tokens": 214422531.0, + "step": 410 + }, + { + "epoch": 0.38658593386120166, + "grad_norm": 0.44827248521714486, + "learning_rate": 4.839489126682776e-05, + "loss": 0.2947, + "num_tokens": 217043971.0, + "step": 415 + }, + { + "epoch": 0.3912435957149511, + "grad_norm": 0.461241958776309, + "learning_rate": 4.830859509837763e-05, + "loss": 0.2871, + "num_tokens": 219639039.0, + "step": 420 + }, + { + "epoch": 0.39590125756870054, + "grad_norm": 0.4360760147983306, + "learning_rate": 4.822229892992751e-05, + "loss": 0.3001, + "num_tokens": 222260479.0, + "step": 425 + }, + { + "epoch": 0.4005589194224499, + "grad_norm": 0.46420621837516945, + "learning_rate": 4.8136002761477395e-05, + "loss": 0.2973, + "num_tokens": 224881919.0, + "step": 430 + }, + { + "epoch": 0.40521658127619936, + "grad_norm": 0.3733620380961576, + "learning_rate": 4.804970659302727e-05, + "loss": 0.2948, + "num_tokens": 227481635.0, + "step": 435 + }, + { + "epoch": 0.40987424312994875, + "grad_norm": 0.4916243745680279, + "learning_rate": 4.796341042457715e-05, + "loss": 0.3006, + "num_tokens": 230103075.0, + "step": 440 + }, + { + "epoch": 0.4145319049836982, + "grad_norm": 0.4402270249594231, + "learning_rate": 4.787711425612703e-05, + "loss": 0.2878, + "num_tokens": 232689331.0, + "step": 445 + }, + { + "epoch": 0.4191895668374476, + "grad_norm": 0.5097510648484884, + "learning_rate": 4.779081808767691e-05, + "loss": 0.2912, + "num_tokens": 235301283.0, + "step": 450 + }, + { + "epoch": 0.423847228691197, + "grad_norm": 0.4733020023413381, + "learning_rate": 4.770452191922679e-05, + "loss": 0.3042, + "num_tokens": 237900010.0, + "step": 455 + }, + { + "epoch": 0.42850489054494645, + "grad_norm": 0.4182733746030012, + "learning_rate": 4.761822575077667e-05, + "loss": 0.2927, + "num_tokens": 240521450.0, + "step": 460 + }, + { + "epoch": 0.43316255239869583, + "grad_norm": 0.433165710595746, + "learning_rate": 4.753192958232654e-05, + "loss": 0.2928, + "num_tokens": 243110250.0, + "step": 465 + }, + { + "epoch": 0.43782021425244527, + "grad_norm": 0.3999472466997859, + "learning_rate": 4.744563341387643e-05, + "loss": 0.2959, + "num_tokens": 245731690.0, + "step": 470 + }, + { + "epoch": 0.4424778761061947, + "grad_norm": 0.4401937042106931, + "learning_rate": 4.7359337245426306e-05, + "loss": 0.3011, + "num_tokens": 248353130.0, + "step": 475 + }, + { + "epoch": 0.4471355379599441, + "grad_norm": 0.39463363710081184, + "learning_rate": 4.7273041076976184e-05, + "loss": 0.295, + "num_tokens": 250974570.0, + "step": 480 + }, + { + "epoch": 0.45179319981369354, + "grad_norm": 0.369259633414095, + "learning_rate": 4.718674490852606e-05, + "loss": 0.2974, + "num_tokens": 253596010.0, + "step": 485 + }, + { + "epoch": 0.4564508616674429, + "grad_norm": 0.42451110897649086, + "learning_rate": 4.710044874007594e-05, + "loss": 0.3037, + "num_tokens": 256216527.0, + "step": 490 + }, + { + "epoch": 0.46110852352119236, + "grad_norm": 0.41192640646628453, + "learning_rate": 4.7014152571625826e-05, + "loss": 0.3106, + "num_tokens": 258829080.0, + "step": 495 + }, + { + "epoch": 0.4657661853749418, + "grad_norm": 0.4565762913092404, + "learning_rate": 4.6927856403175704e-05, + "loss": 0.3355, + "num_tokens": 261450520.0, + "step": 500 + }, + { + "epoch": 0.4704238472286912, + "grad_norm": 0.3565391494715242, + "learning_rate": 4.684156023472558e-05, + "loss": 0.3478, + "num_tokens": 264063793.0, + "step": 505 + }, + { + "epoch": 0.4750815090824406, + "grad_norm": 0.3391571716677179, + "learning_rate": 4.675526406627546e-05, + "loss": 0.3465, + "num_tokens": 266677642.0, + "step": 510 + }, + { + "epoch": 0.47973917093619, + "grad_norm": 0.3752589674406903, + "learning_rate": 4.666896789782534e-05, + "loss": 0.3548, + "num_tokens": 269299082.0, + "step": 515 + }, + { + "epoch": 0.48439683278993945, + "grad_norm": 0.4003513088677163, + "learning_rate": 4.658267172937522e-05, + "loss": 0.3502, + "num_tokens": 271920522.0, + "step": 520 + }, + { + "epoch": 0.4890544946436889, + "grad_norm": 0.37773061821039194, + "learning_rate": 4.64963755609251e-05, + "loss": 0.3572, + "num_tokens": 274541962.0, + "step": 525 + }, + { + "epoch": 0.49371215649743827, + "grad_norm": 0.3578568943439884, + "learning_rate": 4.641007939247497e-05, + "loss": 0.3464, + "num_tokens": 277161966.0, + "step": 530 + }, + { + "epoch": 0.4983698183511877, + "grad_norm": 0.3506179208934016, + "learning_rate": 4.632378322402486e-05, + "loss": 0.3513, + "num_tokens": 279777840.0, + "step": 535 + }, + { + "epoch": 0.5030274802049371, + "grad_norm": 0.3972896159155102, + "learning_rate": 4.6237487055574736e-05, + "loss": 0.3545, + "num_tokens": 282399280.0, + "step": 540 + }, + { + "epoch": 0.5076851420586865, + "grad_norm": 0.4223297300478965, + "learning_rate": 4.6151190887124615e-05, + "loss": 0.345, + "num_tokens": 284959243.0, + "step": 545 + }, + { + "epoch": 0.5123428039124359, + "grad_norm": 0.40153010744650247, + "learning_rate": 4.606489471867449e-05, + "loss": 0.3596, + "num_tokens": 287580360.0, + "step": 550 + }, + { + "epoch": 0.5170004657661854, + "grad_norm": 0.4154707265721631, + "learning_rate": 4.597859855022437e-05, + "loss": 0.3506, + "num_tokens": 290201800.0, + "step": 555 + }, + { + "epoch": 0.5216581276199348, + "grad_norm": 0.4455642719609879, + "learning_rate": 4.589230238177425e-05, + "loss": 0.3528, + "num_tokens": 292817720.0, + "step": 560 + }, + { + "epoch": 0.5263157894736842, + "grad_norm": 0.3997933363319544, + "learning_rate": 4.5806006213324134e-05, + "loss": 0.3519, + "num_tokens": 295439160.0, + "step": 565 + }, + { + "epoch": 0.5309734513274337, + "grad_norm": 0.31094395987853973, + "learning_rate": 4.5719710044874006e-05, + "loss": 0.3508, + "num_tokens": 298031803.0, + "step": 570 + }, + { + "epoch": 0.5356311131811831, + "grad_norm": 0.309211035269928, + "learning_rate": 4.563341387642389e-05, + "loss": 0.3489, + "num_tokens": 300653243.0, + "step": 575 + }, + { + "epoch": 0.5402887750349324, + "grad_norm": 0.359619218023151, + "learning_rate": 4.554711770797377e-05, + "loss": 0.3471, + "num_tokens": 303225503.0, + "step": 580 + }, + { + "epoch": 0.5449464368886818, + "grad_norm": 0.34255420079750104, + "learning_rate": 4.546082153952365e-05, + "loss": 0.3476, + "num_tokens": 305846943.0, + "step": 585 + }, + { + "epoch": 0.5496040987424313, + "grad_norm": 0.40542182146127437, + "learning_rate": 4.5374525371073526e-05, + "loss": 0.3606, + "num_tokens": 308468383.0, + "step": 590 + }, + { + "epoch": 0.5542617605961807, + "grad_norm": 0.4491850870380039, + "learning_rate": 4.5288229202623404e-05, + "loss": 0.3535, + "num_tokens": 311089823.0, + "step": 595 + }, + { + "epoch": 0.5589194224499301, + "grad_norm": 0.4501517095082673, + "learning_rate": 4.520193303417328e-05, + "loss": 0.3552, + "num_tokens": 313711263.0, + "step": 600 + }, + { + "epoch": 0.5635770843036796, + "grad_norm": 0.3635512034769944, + "learning_rate": 4.511563686572317e-05, + "loss": 0.3365, + "num_tokens": 316309964.0, + "step": 605 + }, + { + "epoch": 0.568234746157429, + "grad_norm": 0.34483733371492037, + "learning_rate": 4.5029340697273045e-05, + "loss": 0.3499, + "num_tokens": 318931404.0, + "step": 610 + }, + { + "epoch": 0.5728924080111784, + "grad_norm": 0.3572559367192021, + "learning_rate": 4.4943044528822923e-05, + "loss": 0.3463, + "num_tokens": 321552844.0, + "step": 615 + }, + { + "epoch": 0.5775500698649279, + "grad_norm": 0.3436731405392265, + "learning_rate": 4.48567483603728e-05, + "loss": 0.3508, + "num_tokens": 324174284.0, + "step": 620 + }, + { + "epoch": 0.5822077317186772, + "grad_norm": 0.5303742013030021, + "learning_rate": 4.477045219192268e-05, + "loss": 0.3568, + "num_tokens": 326795724.0, + "step": 625 + }, + { + "epoch": 0.5868653935724266, + "grad_norm": 0.40428231280479165, + "learning_rate": 4.4684156023472565e-05, + "loss": 0.3463, + "num_tokens": 329417164.0, + "step": 630 + }, + { + "epoch": 0.5915230554261761, + "grad_norm": 0.3432839299493264, + "learning_rate": 4.4597859855022436e-05, + "loss": 0.3465, + "num_tokens": 332014390.0, + "step": 635 + }, + { + "epoch": 0.5961807172799255, + "grad_norm": 0.4220791189143345, + "learning_rate": 4.4511563686572315e-05, + "loss": 0.3508, + "num_tokens": 334628125.0, + "step": 640 + }, + { + "epoch": 0.6008383791336749, + "grad_norm": 0.4586942895958512, + "learning_rate": 4.44252675181222e-05, + "loss": 0.3488, + "num_tokens": 337230275.0, + "step": 645 + }, + { + "epoch": 0.6054960409874243, + "grad_norm": 0.3714109798341085, + "learning_rate": 4.433897134967208e-05, + "loss": 0.3499, + "num_tokens": 339850764.0, + "step": 650 + }, + { + "epoch": 0.6101537028411738, + "grad_norm": 0.4008926116406693, + "learning_rate": 4.4252675181221956e-05, + "loss": 0.3541, + "num_tokens": 342472204.0, + "step": 655 + }, + { + "epoch": 0.6148113646949231, + "grad_norm": 0.34765377356863303, + "learning_rate": 4.4166379012771834e-05, + "loss": 0.3543, + "num_tokens": 345093644.0, + "step": 660 + }, + { + "epoch": 0.6194690265486725, + "grad_norm": 0.32332322072308267, + "learning_rate": 4.408008284432171e-05, + "loss": 0.3485, + "num_tokens": 347686320.0, + "step": 665 + }, + { + "epoch": 0.624126688402422, + "grad_norm": 0.35580436954639033, + "learning_rate": 4.39937866758716e-05, + "loss": 0.3524, + "num_tokens": 350307760.0, + "step": 670 + }, + { + "epoch": 0.6287843502561714, + "grad_norm": 0.4333129378651075, + "learning_rate": 4.3907490507421476e-05, + "loss": 0.3479, + "num_tokens": 352915502.0, + "step": 675 + }, + { + "epoch": 0.6334420121099208, + "grad_norm": 0.33390835131596075, + "learning_rate": 4.382119433897135e-05, + "loss": 0.3494, + "num_tokens": 355507604.0, + "step": 680 + }, + { + "epoch": 0.6380996739636703, + "grad_norm": 0.357173329760582, + "learning_rate": 4.373489817052123e-05, + "loss": 0.3576, + "num_tokens": 358102620.0, + "step": 685 + }, + { + "epoch": 0.6427573358174197, + "grad_norm": 0.355905703126643, + "learning_rate": 4.364860200207111e-05, + "loss": 0.3508, + "num_tokens": 360724060.0, + "step": 690 + }, + { + "epoch": 0.6474149976711691, + "grad_norm": 0.3923025321797665, + "learning_rate": 4.356230583362099e-05, + "loss": 0.3512, + "num_tokens": 363345500.0, + "step": 695 + }, + { + "epoch": 0.6520726595249184, + "grad_norm": 0.4615340443716854, + "learning_rate": 4.347600966517087e-05, + "loss": 0.3488, + "num_tokens": 365965454.0, + "step": 700 + }, + { + "epoch": 0.6567303213786679, + "grad_norm": 0.3508078022803716, + "learning_rate": 4.3389713496720745e-05, + "loss": 0.3466, + "num_tokens": 368566578.0, + "step": 705 + }, + { + "epoch": 0.6613879832324173, + "grad_norm": 0.39418653114960156, + "learning_rate": 4.330341732827063e-05, + "loss": 0.3554, + "num_tokens": 371168165.0, + "step": 710 + }, + { + "epoch": 0.6660456450861667, + "grad_norm": 0.5091202912446522, + "learning_rate": 4.321712115982051e-05, + "loss": 0.3512, + "num_tokens": 373789605.0, + "step": 715 + }, + { + "epoch": 0.6707033069399162, + "grad_norm": 0.35769440707050915, + "learning_rate": 4.3130824991370387e-05, + "loss": 0.3475, + "num_tokens": 376376000.0, + "step": 720 + }, + { + "epoch": 0.6753609687936656, + "grad_norm": 0.30703298011124236, + "learning_rate": 4.3044528822920265e-05, + "loss": 0.3529, + "num_tokens": 378997440.0, + "step": 725 + }, + { + "epoch": 0.680018630647415, + "grad_norm": 0.40296087619428983, + "learning_rate": 4.295823265447014e-05, + "loss": 0.3561, + "num_tokens": 381618880.0, + "step": 730 + }, + { + "epoch": 0.6846762925011645, + "grad_norm": 0.3305890378890626, + "learning_rate": 4.287193648602002e-05, + "loss": 0.3451, + "num_tokens": 384240320.0, + "step": 735 + }, + { + "epoch": 0.6893339543549138, + "grad_norm": 0.4068405386758603, + "learning_rate": 4.27856403175699e-05, + "loss": 0.348, + "num_tokens": 386861760.0, + "step": 740 + }, + { + "epoch": 0.6939916162086632, + "grad_norm": 0.3324914070368526, + "learning_rate": 4.269934414911978e-05, + "loss": 0.3484, + "num_tokens": 389483200.0, + "step": 745 + }, + { + "epoch": 0.6986492780624126, + "grad_norm": 0.38594952569803137, + "learning_rate": 4.261304798066966e-05, + "loss": 0.3531, + "num_tokens": 392104640.0, + "step": 750 + }, + { + "epoch": 0.7033069399161621, + "grad_norm": 0.3904186621380532, + "learning_rate": 4.252675181221954e-05, + "loss": 0.3445, + "num_tokens": 394726080.0, + "step": 755 + }, + { + "epoch": 0.7079646017699115, + "grad_norm": 0.34862987122765965, + "learning_rate": 4.244045564376942e-05, + "loss": 0.3456, + "num_tokens": 397347520.0, + "step": 760 + }, + { + "epoch": 0.7126222636236609, + "grad_norm": 0.31776518778099616, + "learning_rate": 4.23541594753193e-05, + "loss": 0.3523, + "num_tokens": 399968838.0, + "step": 765 + }, + { + "epoch": 0.7172799254774104, + "grad_norm": 0.4042167640996357, + "learning_rate": 4.2267863306869176e-05, + "loss": 0.3533, + "num_tokens": 402590278.0, + "step": 770 + }, + { + "epoch": 0.7219375873311598, + "grad_norm": 0.48879019149496866, + "learning_rate": 4.2181567138419054e-05, + "loss": 0.3438, + "num_tokens": 405198917.0, + "step": 775 + }, + { + "epoch": 0.7265952491849091, + "grad_norm": 0.4174350631641972, + "learning_rate": 4.209527096996894e-05, + "loss": 0.3563, + "num_tokens": 407782773.0, + "step": 780 + }, + { + "epoch": 0.7312529110386586, + "grad_norm": 0.2959354186783129, + "learning_rate": 4.200897480151881e-05, + "loss": 0.3518, + "num_tokens": 410401323.0, + "step": 785 + }, + { + "epoch": 0.735910572892408, + "grad_norm": 0.3538958333784061, + "learning_rate": 4.1922678633068695e-05, + "loss": 0.3463, + "num_tokens": 413022763.0, + "step": 790 + }, + { + "epoch": 0.7405682347461574, + "grad_norm": 0.3209486833709415, + "learning_rate": 4.1836382464618573e-05, + "loss": 0.3505, + "num_tokens": 415622538.0, + "step": 795 + }, + { + "epoch": 0.7452258965999069, + "grad_norm": 0.32824211239120443, + "learning_rate": 4.175008629616845e-05, + "loss": 0.3492, + "num_tokens": 418243978.0, + "step": 800 + }, + { + "epoch": 0.7498835584536563, + "grad_norm": 0.3388544675623175, + "learning_rate": 4.166379012771833e-05, + "loss": 0.3596, + "num_tokens": 420835113.0, + "step": 805 + }, + { + "epoch": 0.7545412203074057, + "grad_norm": 0.35049650782126746, + "learning_rate": 4.157749395926821e-05, + "loss": 0.3492, + "num_tokens": 423445640.0, + "step": 810 + }, + { + "epoch": 0.759198882161155, + "grad_norm": 0.43085749480945107, + "learning_rate": 4.1491197790818086e-05, + "loss": 0.3516, + "num_tokens": 426038048.0, + "step": 815 + }, + { + "epoch": 0.7638565440149045, + "grad_norm": 0.3800016220245255, + "learning_rate": 4.140490162236797e-05, + "loss": 0.3536, + "num_tokens": 428659488.0, + "step": 820 + }, + { + "epoch": 0.7685142058686539, + "grad_norm": 0.40534919951359066, + "learning_rate": 4.131860545391785e-05, + "loss": 0.3458, + "num_tokens": 431280928.0, + "step": 825 + }, + { + "epoch": 0.7731718677224033, + "grad_norm": 0.3282502514421546, + "learning_rate": 4.123230928546773e-05, + "loss": 0.351, + "num_tokens": 433902368.0, + "step": 830 + }, + { + "epoch": 0.7778295295761528, + "grad_norm": 0.3425108014869088, + "learning_rate": 4.1146013117017606e-05, + "loss": 0.3564, + "num_tokens": 436522664.0, + "step": 835 + }, + { + "epoch": 0.7824871914299022, + "grad_norm": 0.33361835217876357, + "learning_rate": 4.1059716948567484e-05, + "loss": 0.3433, + "num_tokens": 439124783.0, + "step": 840 + }, + { + "epoch": 0.7871448532836516, + "grad_norm": 0.33798159635466163, + "learning_rate": 4.097342078011737e-05, + "loss": 0.3496, + "num_tokens": 441746223.0, + "step": 845 + }, + { + "epoch": 0.7918025151374011, + "grad_norm": 0.447347998974799, + "learning_rate": 4.088712461166724e-05, + "loss": 0.351, + "num_tokens": 444367663.0, + "step": 850 + }, + { + "epoch": 0.7964601769911505, + "grad_norm": 0.3525745631003211, + "learning_rate": 4.080082844321712e-05, + "loss": 0.3457, + "num_tokens": 446989103.0, + "step": 855 + }, + { + "epoch": 0.8011178388448998, + "grad_norm": 0.31916567647993294, + "learning_rate": 4.0714532274767004e-05, + "loss": 0.3451, + "num_tokens": 449590185.0, + "step": 860 + }, + { + "epoch": 0.8057755006986492, + "grad_norm": 0.318983153886092, + "learning_rate": 4.062823610631688e-05, + "loss": 0.3485, + "num_tokens": 452202740.0, + "step": 865 + }, + { + "epoch": 0.8104331625523987, + "grad_norm": 0.29667088926699325, + "learning_rate": 4.054193993786676e-05, + "loss": 0.3444, + "num_tokens": 454824180.0, + "step": 870 + }, + { + "epoch": 0.8150908244061481, + "grad_norm": 0.31207562700596997, + "learning_rate": 4.045564376941664e-05, + "loss": 0.3517, + "num_tokens": 457445620.0, + "step": 875 + }, + { + "epoch": 0.8197484862598975, + "grad_norm": 0.34870985935788257, + "learning_rate": 4.036934760096652e-05, + "loss": 0.3394, + "num_tokens": 460067060.0, + "step": 880 + }, + { + "epoch": 0.824406148113647, + "grad_norm": 0.30388580216897826, + "learning_rate": 4.02830514325164e-05, + "loss": 0.3507, + "num_tokens": 462674172.0, + "step": 885 + }, + { + "epoch": 0.8290638099673964, + "grad_norm": 0.32472261917860645, + "learning_rate": 4.019675526406628e-05, + "loss": 0.3501, + "num_tokens": 465295612.0, + "step": 890 + }, + { + "epoch": 0.8337214718211458, + "grad_norm": 0.3285171952667728, + "learning_rate": 4.011045909561615e-05, + "loss": 0.3504, + "num_tokens": 467917052.0, + "step": 895 + }, + { + "epoch": 0.8383791336748952, + "grad_norm": 0.3323900078743967, + "learning_rate": 4.0024162927166037e-05, + "loss": 0.3471, + "num_tokens": 470538492.0, + "step": 900 + }, + { + "epoch": 0.8430367955286446, + "grad_norm": 0.32807804825654646, + "learning_rate": 3.9937866758715915e-05, + "loss": 0.3511, + "num_tokens": 473159932.0, + "step": 905 + }, + { + "epoch": 0.847694457382394, + "grad_norm": 0.28919453745593604, + "learning_rate": 3.98515705902658e-05, + "loss": 0.3545, + "num_tokens": 475757711.0, + "step": 910 + }, + { + "epoch": 0.8523521192361434, + "grad_norm": 0.46516583490606656, + "learning_rate": 3.976527442181567e-05, + "loss": 0.3487, + "num_tokens": 478352124.0, + "step": 915 + }, + { + "epoch": 0.8570097810898929, + "grad_norm": 0.41404074008661385, + "learning_rate": 3.967897825336555e-05, + "loss": 0.3544, + "num_tokens": 480973564.0, + "step": 920 + }, + { + "epoch": 0.8616674429436423, + "grad_norm": 0.363700040967719, + "learning_rate": 3.9592682084915434e-05, + "loss": 0.35, + "num_tokens": 483595004.0, + "step": 925 + }, + { + "epoch": 0.8663251047973917, + "grad_norm": 0.42236484445152317, + "learning_rate": 3.950638591646531e-05, + "loss": 0.3531, + "num_tokens": 486216444.0, + "step": 930 + }, + { + "epoch": 0.8709827666511412, + "grad_norm": 0.4078726261905896, + "learning_rate": 3.942008974801519e-05, + "loss": 0.3411, + "num_tokens": 488837884.0, + "step": 935 + }, + { + "epoch": 0.8756404285048905, + "grad_norm": 0.3918480578581464, + "learning_rate": 3.933379357956507e-05, + "loss": 0.3441, + "num_tokens": 491454413.0, + "step": 940 + }, + { + "epoch": 0.8802980903586399, + "grad_norm": 0.3170963095548561, + "learning_rate": 3.924749741111495e-05, + "loss": 0.3456, + "num_tokens": 494064784.0, + "step": 945 + }, + { + "epoch": 0.8849557522123894, + "grad_norm": 0.3208860567524943, + "learning_rate": 3.916120124266483e-05, + "loss": 0.338, + "num_tokens": 496686224.0, + "step": 950 + }, + { + "epoch": 0.8896134140661388, + "grad_norm": 0.3597661575787414, + "learning_rate": 3.9074905074214704e-05, + "loss": 0.3494, + "num_tokens": 499307664.0, + "step": 955 + }, + { + "epoch": 0.8942710759198882, + "grad_norm": 0.43417260548610914, + "learning_rate": 3.898860890576458e-05, + "loss": 0.3528, + "num_tokens": 501920213.0, + "step": 960 + }, + { + "epoch": 0.8989287377736377, + "grad_norm": 0.31106495064645334, + "learning_rate": 3.890231273731447e-05, + "loss": 0.3498, + "num_tokens": 504541653.0, + "step": 965 + }, + { + "epoch": 0.9035863996273871, + "grad_norm": 0.35221034421913827, + "learning_rate": 3.8816016568864345e-05, + "loss": 0.353, + "num_tokens": 507163093.0, + "step": 970 + }, + { + "epoch": 0.9082440614811365, + "grad_norm": 0.33006907184172946, + "learning_rate": 3.8729720400414224e-05, + "loss": 0.3495, + "num_tokens": 509784533.0, + "step": 975 + }, + { + "epoch": 0.9129017233348858, + "grad_norm": 0.463640975856773, + "learning_rate": 3.86434242319641e-05, + "loss": 0.3554, + "num_tokens": 512405973.0, + "step": 980 + }, + { + "epoch": 0.9175593851886353, + "grad_norm": 0.27816047488495105, + "learning_rate": 3.855712806351398e-05, + "loss": 0.3631, + "num_tokens": 515027413.0, + "step": 985 + }, + { + "epoch": 0.9222170470423847, + "grad_norm": 2.168338544722683, + "learning_rate": 3.8470831895063865e-05, + "loss": 0.3456, + "num_tokens": 517639572.0, + "step": 990 + }, + { + "epoch": 0.9268747088961341, + "grad_norm": 0.38328693899390504, + "learning_rate": 3.838453572661374e-05, + "loss": 0.3711, + "num_tokens": 520261012.0, + "step": 995 + }, + { + "epoch": 0.9315323707498836, + "grad_norm": 0.4244503538190939, + "learning_rate": 3.8298239558163615e-05, + "loss": 0.3914, + "num_tokens": 522882452.0, + "step": 1000 + }, + { + "epoch": 0.936190032603633, + "grad_norm": 0.591471122073177, + "learning_rate": 3.82119433897135e-05, + "loss": 0.4168, + "num_tokens": 525503892.0, + "step": 1005 + }, + { + "epoch": 0.9408476944573824, + "grad_norm": 0.37635159444985006, + "learning_rate": 3.812564722126338e-05, + "loss": 0.407, + "num_tokens": 528125332.0, + "step": 1010 + }, + { + "epoch": 0.9455053563111319, + "grad_norm": 0.41163572473058424, + "learning_rate": 3.8039351052813256e-05, + "loss": 0.4084, + "num_tokens": 530746772.0, + "step": 1015 + }, + { + "epoch": 0.9501630181648812, + "grad_norm": 0.3843982365118871, + "learning_rate": 3.7953054884363134e-05, + "loss": 0.4162, + "num_tokens": 533368212.0, + "step": 1020 + }, + { + "epoch": 0.9548206800186306, + "grad_norm": 0.4651466028200984, + "learning_rate": 3.786675871591301e-05, + "loss": 0.4091, + "num_tokens": 535989652.0, + "step": 1025 + }, + { + "epoch": 0.95947834187238, + "grad_norm": 0.32589812875265384, + "learning_rate": 3.77804625474629e-05, + "loss": 0.4003, + "num_tokens": 538611092.0, + "step": 1030 + }, + { + "epoch": 0.9641360037261295, + "grad_norm": 0.3671450936074502, + "learning_rate": 3.7694166379012776e-05, + "loss": 0.4032, + "num_tokens": 541232532.0, + "step": 1035 + }, + { + "epoch": 0.9687936655798789, + "grad_norm": 0.30799336780124387, + "learning_rate": 3.7607870210562654e-05, + "loss": 0.403, + "num_tokens": 543853972.0, + "step": 1040 + }, + { + "epoch": 0.9734513274336283, + "grad_norm": 0.2683570971972702, + "learning_rate": 3.752157404211253e-05, + "loss": 0.4033, + "num_tokens": 546461523.0, + "step": 1045 + }, + { + "epoch": 0.9781089892873778, + "grad_norm": 0.3238139917000379, + "learning_rate": 3.743527787366241e-05, + "loss": 0.4027, + "num_tokens": 549082963.0, + "step": 1050 + }, + { + "epoch": 0.9827666511411272, + "grad_norm": 0.4169093587901815, + "learning_rate": 3.734898170521229e-05, + "loss": 0.4075, + "num_tokens": 551662348.0, + "step": 1055 + }, + { + "epoch": 0.9874243129948765, + "grad_norm": 0.29918767290268117, + "learning_rate": 3.7262685536762174e-05, + "loss": 0.4062, + "num_tokens": 554283788.0, + "step": 1060 + }, + { + "epoch": 0.992081974848626, + "grad_norm": 0.2831248421544475, + "learning_rate": 3.7176389368312045e-05, + "loss": 0.4042, + "num_tokens": 556905228.0, + "step": 1065 + }, + { + "epoch": 0.9967396367023754, + "grad_norm": 0.322460754267741, + "learning_rate": 3.709009319986193e-05, + "loss": 0.3957, + "num_tokens": 559526668.0, + "step": 1070 + }, + { + "epoch": 1.00093153237075, + "grad_norm": 0.5158478203711624, + "learning_rate": 3.700379703141181e-05, + "loss": 0.3987, + "num_tokens": 561722124.0, + "step": 1075 + }, + { + "epoch": 1.0055891942244992, + "grad_norm": 0.3842474431873255, + "learning_rate": 3.6917500862961687e-05, + "loss": 0.2967, + "num_tokens": 564312816.0, + "step": 1080 + }, + { + "epoch": 1.0102468560782487, + "grad_norm": 0.30902058029027224, + "learning_rate": 3.6831204694511565e-05, + "loss": 0.2768, + "num_tokens": 566934256.0, + "step": 1085 + }, + { + "epoch": 1.0149045179319982, + "grad_norm": 0.27843760124356054, + "learning_rate": 3.674490852606144e-05, + "loss": 0.2817, + "num_tokens": 569555696.0, + "step": 1090 + }, + { + "epoch": 1.0195621797857475, + "grad_norm": 0.29052535529866214, + "learning_rate": 3.665861235761132e-05, + "loss": 0.276, + "num_tokens": 572156684.0, + "step": 1095 + }, + { + "epoch": 1.024219841639497, + "grad_norm": 0.27630310315466183, + "learning_rate": 3.6572316189161206e-05, + "loss": 0.2797, + "num_tokens": 574749092.0, + "step": 1100 + }, + { + "epoch": 1.0288775034932465, + "grad_norm": 0.278687574814992, + "learning_rate": 3.6486020020711085e-05, + "loss": 0.2787, + "num_tokens": 577370532.0, + "step": 1105 + }, + { + "epoch": 1.0335351653469957, + "grad_norm": 0.23131688150393118, + "learning_rate": 3.639972385226096e-05, + "loss": 0.2818, + "num_tokens": 579991972.0, + "step": 1110 + }, + { + "epoch": 1.0381928272007452, + "grad_norm": 0.29831078797973526, + "learning_rate": 3.631342768381084e-05, + "loss": 0.2685, + "num_tokens": 582591688.0, + "step": 1115 + }, + { + "epoch": 1.0428504890544947, + "grad_norm": 0.2527756012565954, + "learning_rate": 3.622713151536072e-05, + "loss": 0.2788, + "num_tokens": 585213128.0, + "step": 1120 + }, + { + "epoch": 1.047508150908244, + "grad_norm": 0.25154350991263563, + "learning_rate": 3.6140835346910604e-05, + "loss": 0.2793, + "num_tokens": 587834568.0, + "step": 1125 + }, + { + "epoch": 1.0521658127619935, + "grad_norm": 0.2850095798638657, + "learning_rate": 3.6054539178460476e-05, + "loss": 0.2657, + "num_tokens": 590415436.0, + "step": 1130 + }, + { + "epoch": 1.056823474615743, + "grad_norm": 0.31180404727290206, + "learning_rate": 3.5968243010010354e-05, + "loss": 0.276, + "num_tokens": 593036876.0, + "step": 1135 + }, + { + "epoch": 1.0614811364694923, + "grad_norm": 0.3349544289972451, + "learning_rate": 3.588194684156024e-05, + "loss": 0.2783, + "num_tokens": 595658316.0, + "step": 1140 + }, + { + "epoch": 1.0661387983232418, + "grad_norm": 0.28515849317064546, + "learning_rate": 3.579565067311012e-05, + "loss": 0.2747, + "num_tokens": 598248306.0, + "step": 1145 + }, + { + "epoch": 1.0707964601769913, + "grad_norm": 0.2558473515039951, + "learning_rate": 3.5709354504659995e-05, + "loss": 0.2783, + "num_tokens": 600869746.0, + "step": 1150 + }, + { + "epoch": 1.0754541220307405, + "grad_norm": 0.3055750325974408, + "learning_rate": 3.5623058336209874e-05, + "loss": 0.2827, + "num_tokens": 603469740.0, + "step": 1155 + }, + { + "epoch": 1.08011178388449, + "grad_norm": 0.2523023068708641, + "learning_rate": 3.553676216775975e-05, + "loss": 0.2915, + "num_tokens": 606032296.0, + "step": 1160 + }, + { + "epoch": 1.0847694457382393, + "grad_norm": 0.26982046820141947, + "learning_rate": 3.545046599930964e-05, + "loss": 0.2833, + "num_tokens": 608640980.0, + "step": 1165 + }, + { + "epoch": 1.0894271075919888, + "grad_norm": 0.29368512739305636, + "learning_rate": 3.536416983085951e-05, + "loss": 0.2795, + "num_tokens": 611262420.0, + "step": 1170 + }, + { + "epoch": 1.0940847694457383, + "grad_norm": 0.3772826473034449, + "learning_rate": 3.5277873662409386e-05, + "loss": 0.2804, + "num_tokens": 613883860.0, + "step": 1175 + }, + { + "epoch": 1.0987424312994876, + "grad_norm": 0.3154485480529681, + "learning_rate": 3.519157749395927e-05, + "loss": 0.2735, + "num_tokens": 616486010.0, + "step": 1180 + }, + { + "epoch": 1.103400093153237, + "grad_norm": 0.26976291529898727, + "learning_rate": 3.510528132550915e-05, + "loss": 0.2769, + "num_tokens": 619107450.0, + "step": 1185 + }, + { + "epoch": 1.1080577550069866, + "grad_norm": 0.2773634956709149, + "learning_rate": 3.501898515705903e-05, + "loss": 0.2616, + "num_tokens": 621728890.0, + "step": 1190 + }, + { + "epoch": 1.1127154168607358, + "grad_norm": 0.30021623861010804, + "learning_rate": 3.4932688988608906e-05, + "loss": 0.2706, + "num_tokens": 624309344.0, + "step": 1195 + }, + { + "epoch": 1.1173730787144853, + "grad_norm": 0.2609207885474365, + "learning_rate": 3.4846392820158784e-05, + "loss": 0.2864, + "num_tokens": 626919167.0, + "step": 1200 + }, + { + "epoch": 1.1220307405682348, + "grad_norm": 0.33890867374743183, + "learning_rate": 3.476009665170867e-05, + "loss": 0.2857, + "num_tokens": 629531326.0, + "step": 1205 + }, + { + "epoch": 1.126688402421984, + "grad_norm": 0.33533079561986484, + "learning_rate": 3.467380048325855e-05, + "loss": 0.2757, + "num_tokens": 632135386.0, + "step": 1210 + }, + { + "epoch": 1.1313460642757336, + "grad_norm": 0.2789080559095753, + "learning_rate": 3.458750431480842e-05, + "loss": 0.2699, + "num_tokens": 634747941.0, + "step": 1215 + }, + { + "epoch": 1.136003726129483, + "grad_norm": 0.2698082442225666, + "learning_rate": 3.4501208146358304e-05, + "loss": 0.282, + "num_tokens": 637344492.0, + "step": 1220 + }, + { + "epoch": 1.1406613879832324, + "grad_norm": 0.29152261058658296, + "learning_rate": 3.441491197790818e-05, + "loss": 0.2673, + "num_tokens": 639965932.0, + "step": 1225 + }, + { + "epoch": 1.1453190498369819, + "grad_norm": 0.26522717303501175, + "learning_rate": 3.432861580945806e-05, + "loss": 0.2757, + "num_tokens": 642587372.0, + "step": 1230 + }, + { + "epoch": 1.1499767116907313, + "grad_norm": 0.2901549255370733, + "learning_rate": 3.424231964100794e-05, + "loss": 0.2728, + "num_tokens": 645178768.0, + "step": 1235 + }, + { + "epoch": 1.1546343735444806, + "grad_norm": 0.3068102140084383, + "learning_rate": 3.415602347255782e-05, + "loss": 0.2786, + "num_tokens": 647793774.0, + "step": 1240 + }, + { + "epoch": 1.1592920353982301, + "grad_norm": 0.2370803484062004, + "learning_rate": 3.40697273041077e-05, + "loss": 0.2884, + "num_tokens": 650415214.0, + "step": 1245 + }, + { + "epoch": 1.1639496972519794, + "grad_norm": 0.28521208376945334, + "learning_rate": 3.398343113565758e-05, + "loss": 0.2807, + "num_tokens": 653036654.0, + "step": 1250 + }, + { + "epoch": 1.1686073591057289, + "grad_norm": 0.26932885849381305, + "learning_rate": 3.389713496720746e-05, + "loss": 0.2763, + "num_tokens": 655624923.0, + "step": 1255 + }, + { + "epoch": 1.1732650209594784, + "grad_norm": 0.2538805120966557, + "learning_rate": 3.381083879875734e-05, + "loss": 0.2806, + "num_tokens": 658246363.0, + "step": 1260 + }, + { + "epoch": 1.1779226828132279, + "grad_norm": 0.243785221427234, + "learning_rate": 3.3724542630307215e-05, + "loss": 0.2792, + "num_tokens": 660864764.0, + "step": 1265 + }, + { + "epoch": 1.1825803446669771, + "grad_norm": 0.35726148914420935, + "learning_rate": 3.363824646185709e-05, + "loss": 0.2695, + "num_tokens": 663461223.0, + "step": 1270 + }, + { + "epoch": 1.1872380065207266, + "grad_norm": 0.3062285105387582, + "learning_rate": 3.355195029340698e-05, + "loss": 0.2862, + "num_tokens": 666082663.0, + "step": 1275 + }, + { + "epoch": 1.191895668374476, + "grad_norm": 0.2965179427739923, + "learning_rate": 3.346565412495685e-05, + "loss": 0.2743, + "num_tokens": 668704103.0, + "step": 1280 + }, + { + "epoch": 1.1965533302282254, + "grad_norm": 0.31841050941054977, + "learning_rate": 3.3379357956506735e-05, + "loss": 0.2852, + "num_tokens": 671325543.0, + "step": 1285 + }, + { + "epoch": 1.201210992081975, + "grad_norm": 0.2668801197416085, + "learning_rate": 3.329306178805661e-05, + "loss": 0.2726, + "num_tokens": 673938014.0, + "step": 1290 + }, + { + "epoch": 1.2058686539357242, + "grad_norm": 0.2619716412853699, + "learning_rate": 3.320676561960649e-05, + "loss": 0.2832, + "num_tokens": 676559454.0, + "step": 1295 + }, + { + "epoch": 1.2105263157894737, + "grad_norm": 0.27075658128215524, + "learning_rate": 3.312046945115637e-05, + "loss": 0.2796, + "num_tokens": 679180894.0, + "step": 1300 + }, + { + "epoch": 1.2151839776432232, + "grad_norm": 0.3034321823924177, + "learning_rate": 3.303417328270625e-05, + "loss": 0.2901, + "num_tokens": 681778120.0, + "step": 1305 + }, + { + "epoch": 1.2198416394969724, + "grad_norm": 0.28559797634175, + "learning_rate": 3.2947877114256126e-05, + "loss": 0.2865, + "num_tokens": 684399560.0, + "step": 1310 + }, + { + "epoch": 1.224499301350722, + "grad_norm": 0.2620025012126218, + "learning_rate": 3.286158094580601e-05, + "loss": 0.2764, + "num_tokens": 687021000.0, + "step": 1315 + }, + { + "epoch": 1.2291569632044714, + "grad_norm": 0.3065365336854166, + "learning_rate": 3.277528477735589e-05, + "loss": 0.283, + "num_tokens": 689642440.0, + "step": 1320 + }, + { + "epoch": 1.2338146250582207, + "grad_norm": 0.31061841933478607, + "learning_rate": 3.268898860890577e-05, + "loss": 0.2865, + "num_tokens": 692242175.0, + "step": 1325 + }, + { + "epoch": 1.2384722869119702, + "grad_norm": 0.30651408147568526, + "learning_rate": 3.2602692440455645e-05, + "loss": 0.2836, + "num_tokens": 694863615.0, + "step": 1330 + }, + { + "epoch": 1.2431299487657197, + "grad_norm": 0.31688698207820437, + "learning_rate": 3.2516396272005524e-05, + "loss": 0.2778, + "num_tokens": 697459919.0, + "step": 1335 + }, + { + "epoch": 1.247787610619469, + "grad_norm": 0.31942806223486203, + "learning_rate": 3.243010010355541e-05, + "loss": 0.2799, + "num_tokens": 700081359.0, + "step": 1340 + }, + { + "epoch": 1.2524452724732185, + "grad_norm": 0.31353101869290123, + "learning_rate": 3.234380393510528e-05, + "loss": 0.2657, + "num_tokens": 702689027.0, + "step": 1345 + }, + { + "epoch": 1.257102934326968, + "grad_norm": 0.32712994813622637, + "learning_rate": 3.225750776665516e-05, + "loss": 0.2824, + "num_tokens": 705310467.0, + "step": 1350 + }, + { + "epoch": 1.2617605961807172, + "grad_norm": 0.25649502957591136, + "learning_rate": 3.217121159820504e-05, + "loss": 0.275, + "num_tokens": 707931907.0, + "step": 1355 + }, + { + "epoch": 1.2664182580344667, + "grad_norm": 0.27787099637497753, + "learning_rate": 3.208491542975492e-05, + "loss": 0.2746, + "num_tokens": 710553347.0, + "step": 1360 + }, + { + "epoch": 1.271075919888216, + "grad_norm": 0.28878789150868767, + "learning_rate": 3.19986192613048e-05, + "loss": 0.2792, + "num_tokens": 713174787.0, + "step": 1365 + }, + { + "epoch": 1.2757335817419655, + "grad_norm": 0.257307550138807, + "learning_rate": 3.191232309285468e-05, + "loss": 0.2801, + "num_tokens": 715796227.0, + "step": 1370 + }, + { + "epoch": 1.280391243595715, + "grad_norm": 0.23609227989037512, + "learning_rate": 3.1826026924404556e-05, + "loss": 0.2793, + "num_tokens": 718417667.0, + "step": 1375 + }, + { + "epoch": 1.2850489054494645, + "grad_norm": 0.2463031753879305, + "learning_rate": 3.173973075595444e-05, + "loss": 0.2859, + "num_tokens": 721039107.0, + "step": 1380 + }, + { + "epoch": 1.2897065673032138, + "grad_norm": 0.2722779089462647, + "learning_rate": 3.165343458750431e-05, + "loss": 0.2766, + "num_tokens": 723660547.0, + "step": 1385 + }, + { + "epoch": 1.2943642291569633, + "grad_norm": 0.3178746369585837, + "learning_rate": 3.156713841905419e-05, + "loss": 0.2766, + "num_tokens": 726281987.0, + "step": 1390 + }, + { + "epoch": 1.2990218910107125, + "grad_norm": 0.30536089345920436, + "learning_rate": 3.1480842250604076e-05, + "loss": 0.2929, + "num_tokens": 728895260.0, + "step": 1395 + }, + { + "epoch": 1.303679552864462, + "grad_norm": 0.323563907379548, + "learning_rate": 3.1394546082153954e-05, + "loss": 0.2757, + "num_tokens": 731507815.0, + "step": 1400 + }, + { + "epoch": 1.3083372147182115, + "grad_norm": 0.29136024043207664, + "learning_rate": 3.130824991370383e-05, + "loss": 0.2691, + "num_tokens": 734108897.0, + "step": 1405 + }, + { + "epoch": 1.312994876571961, + "grad_norm": 0.29970041422727933, + "learning_rate": 3.122195374525371e-05, + "loss": 0.2798, + "num_tokens": 736686494.0, + "step": 1410 + }, + { + "epoch": 1.3176525384257103, + "grad_norm": 0.26717219505855055, + "learning_rate": 3.113565757680359e-05, + "loss": 0.2963, + "num_tokens": 739305242.0, + "step": 1415 + }, + { + "epoch": 1.3223102002794598, + "grad_norm": 0.30166706406997973, + "learning_rate": 3.1049361408353474e-05, + "loss": 0.2911, + "num_tokens": 741926682.0, + "step": 1420 + }, + { + "epoch": 1.326967862133209, + "grad_norm": 0.23037762741268314, + "learning_rate": 3.096306523990335e-05, + "loss": 0.2688, + "num_tokens": 744521095.0, + "step": 1425 + }, + { + "epoch": 1.3316255239869585, + "grad_norm": 0.2513818523895221, + "learning_rate": 3.0876769071453223e-05, + "loss": 0.2822, + "num_tokens": 747116111.0, + "step": 1430 + }, + { + "epoch": 1.336283185840708, + "grad_norm": 0.25728555675981446, + "learning_rate": 3.079047290300311e-05, + "loss": 0.2833, + "num_tokens": 749737551.0, + "step": 1435 + }, + { + "epoch": 1.3409408476944573, + "grad_norm": 0.2564221870409054, + "learning_rate": 3.070417673455299e-05, + "loss": 0.2808, + "num_tokens": 752358040.0, + "step": 1440 + }, + { + "epoch": 1.3455985095482068, + "grad_norm": 0.254766793549684, + "learning_rate": 3.061788056610287e-05, + "loss": 0.285, + "num_tokens": 754979480.0, + "step": 1445 + }, + { + "epoch": 1.350256171401956, + "grad_norm": 0.24498900221791703, + "learning_rate": 3.053158439765274e-05, + "loss": 0.2754, + "num_tokens": 757600920.0, + "step": 1450 + }, + { + "epoch": 1.3549138332557056, + "grad_norm": 0.23493264294585314, + "learning_rate": 3.0445288229202625e-05, + "loss": 0.2814, + "num_tokens": 760222360.0, + "step": 1455 + }, + { + "epoch": 1.359571495109455, + "grad_norm": 0.2630386608386504, + "learning_rate": 3.0358992060752506e-05, + "loss": 0.2637, + "num_tokens": 762843678.0, + "step": 1460 + }, + { + "epoch": 1.3642291569632046, + "grad_norm": 0.2394316260345572, + "learning_rate": 3.027269589230238e-05, + "loss": 0.2796, + "num_tokens": 765442631.0, + "step": 1465 + }, + { + "epoch": 1.3688868188169538, + "grad_norm": 0.2390476917151645, + "learning_rate": 3.018639972385226e-05, + "loss": 0.2814, + "num_tokens": 768049921.0, + "step": 1470 + }, + { + "epoch": 1.3735444806707033, + "grad_norm": 0.2919140428129752, + "learning_rate": 3.010010355540214e-05, + "loss": 0.2772, + "num_tokens": 770646930.0, + "step": 1475 + }, + { + "epoch": 1.3782021425244526, + "grad_norm": 0.29385816246884944, + "learning_rate": 3.001380738695202e-05, + "loss": 0.2884, + "num_tokens": 773240766.0, + "step": 1480 + }, + { + "epoch": 1.382859804378202, + "grad_norm": 0.33487740297263696, + "learning_rate": 2.99275112185019e-05, + "loss": 0.2908, + "num_tokens": 775862206.0, + "step": 1485 + }, + { + "epoch": 1.3875174662319516, + "grad_norm": 0.2577851937226835, + "learning_rate": 2.984121505005178e-05, + "loss": 0.2867, + "num_tokens": 778483646.0, + "step": 1490 + }, + { + "epoch": 1.392175128085701, + "grad_norm": 0.27996717526951814, + "learning_rate": 2.9754918881601657e-05, + "loss": 0.2791, + "num_tokens": 781105086.0, + "step": 1495 + }, + { + "epoch": 1.3968327899394504, + "grad_norm": 0.25754838201668395, + "learning_rate": 2.966862271315154e-05, + "loss": 0.2686, + "num_tokens": 783726526.0, + "step": 1500 + }, + { + "epoch": 1.4014904517931999, + "grad_norm": 0.300697823170609, + "learning_rate": 2.9582326544701417e-05, + "loss": 0.2766, + "num_tokens": 786325616.0, + "step": 1505 + }, + { + "epoch": 1.4061481136469491, + "grad_norm": 0.35724755171638073, + "learning_rate": 2.9496030376251292e-05, + "loss": 0.2724, + "num_tokens": 788947056.0, + "step": 1510 + }, + { + "epoch": 1.4108057755006986, + "grad_norm": 0.24677399714229475, + "learning_rate": 2.9409734207801177e-05, + "loss": 0.2816, + "num_tokens": 791538191.0, + "step": 1515 + }, + { + "epoch": 1.4154634373544481, + "grad_norm": 0.26435033183990186, + "learning_rate": 2.9323438039351052e-05, + "loss": 0.2734, + "num_tokens": 794159631.0, + "step": 1520 + }, + { + "epoch": 1.4201210992081974, + "grad_norm": 0.32036106613904486, + "learning_rate": 2.9237141870900937e-05, + "loss": 0.2819, + "num_tokens": 796781071.0, + "step": 1525 + }, + { + "epoch": 1.424778761061947, + "grad_norm": 0.2475476137169364, + "learning_rate": 2.9150845702450812e-05, + "loss": 0.2688, + "num_tokens": 799402511.0, + "step": 1530 + }, + { + "epoch": 1.4294364229156964, + "grad_norm": 0.2528684630121611, + "learning_rate": 2.906454953400069e-05, + "loss": 0.2883, + "num_tokens": 802023951.0, + "step": 1535 + }, + { + "epoch": 1.4340940847694457, + "grad_norm": 0.22968741877065332, + "learning_rate": 2.897825336555057e-05, + "loss": 0.2746, + "num_tokens": 804645391.0, + "step": 1540 + }, + { + "epoch": 1.4387517466231952, + "grad_norm": 0.24342123000808089, + "learning_rate": 2.889195719710045e-05, + "loss": 0.271, + "num_tokens": 807266831.0, + "step": 1545 + }, + { + "epoch": 1.4434094084769447, + "grad_norm": 0.29535237660947017, + "learning_rate": 2.8805661028650328e-05, + "loss": 0.2692, + "num_tokens": 809888271.0, + "step": 1550 + }, + { + "epoch": 1.448067070330694, + "grad_norm": 0.275059474845138, + "learning_rate": 2.871936486020021e-05, + "loss": 0.2689, + "num_tokens": 812509711.0, + "step": 1555 + }, + { + "epoch": 1.4527247321844434, + "grad_norm": 0.26490502059184046, + "learning_rate": 2.8633068691750088e-05, + "loss": 0.28, + "num_tokens": 815131151.0, + "step": 1560 + }, + { + "epoch": 1.4573823940381927, + "grad_norm": 0.23510205220169592, + "learning_rate": 2.854677252329997e-05, + "loss": 0.2858, + "num_tokens": 817730759.0, + "step": 1565 + }, + { + "epoch": 1.4620400558919422, + "grad_norm": 0.3374158150586498, + "learning_rate": 2.8460476354849848e-05, + "loss": 0.2978, + "num_tokens": 820352199.0, + "step": 1570 + }, + { + "epoch": 1.4666977177456917, + "grad_norm": 0.26454799397527445, + "learning_rate": 2.8374180186399723e-05, + "loss": 0.2737, + "num_tokens": 822973639.0, + "step": 1575 + }, + { + "epoch": 1.4713553795994412, + "grad_norm": 0.26762807423610585, + "learning_rate": 2.8287884017949608e-05, + "loss": 0.2782, + "num_tokens": 825595079.0, + "step": 1580 + }, + { + "epoch": 1.4760130414531905, + "grad_norm": 0.24025946078804356, + "learning_rate": 2.8201587849499482e-05, + "loss": 0.2795, + "num_tokens": 828216519.0, + "step": 1585 + }, + { + "epoch": 1.48067070330694, + "grad_norm": 0.24596937742327113, + "learning_rate": 2.811529168104936e-05, + "loss": 0.2768, + "num_tokens": 830837959.0, + "step": 1590 + }, + { + "epoch": 1.4853283651606892, + "grad_norm": 0.2356611967272471, + "learning_rate": 2.8028995512599242e-05, + "loss": 0.2786, + "num_tokens": 833459399.0, + "step": 1595 + }, + { + "epoch": 1.4899860270144387, + "grad_norm": 0.24597105088414528, + "learning_rate": 2.794269934414912e-05, + "loss": 0.286, + "num_tokens": 836080839.0, + "step": 1600 + }, + { + "epoch": 1.4946436888681882, + "grad_norm": 0.2413954227984674, + "learning_rate": 2.7856403175699002e-05, + "loss": 0.2829, + "num_tokens": 838689704.0, + "step": 1605 + }, + { + "epoch": 1.4993013507219377, + "grad_norm": 0.25744774684417754, + "learning_rate": 2.777010700724888e-05, + "loss": 0.2793, + "num_tokens": 841287483.0, + "step": 1610 + }, + { + "epoch": 1.503959012575687, + "grad_norm": 0.2821271289814806, + "learning_rate": 2.768381083879876e-05, + "loss": 0.2776, + "num_tokens": 843905126.0, + "step": 1615 + }, + { + "epoch": 1.5086166744294365, + "grad_norm": 0.29639194566235333, + "learning_rate": 2.759751467034864e-05, + "loss": 0.2888, + "num_tokens": 846525643.0, + "step": 1620 + }, + { + "epoch": 1.5132743362831858, + "grad_norm": 0.24055196328513787, + "learning_rate": 2.751121850189852e-05, + "loss": 0.2859, + "num_tokens": 849147083.0, + "step": 1625 + }, + { + "epoch": 1.5179319981369352, + "grad_norm": 0.2624395571193606, + "learning_rate": 2.7424922333448393e-05, + "loss": 0.2817, + "num_tokens": 851768523.0, + "step": 1630 + }, + { + "epoch": 1.5225896599906847, + "grad_norm": 0.2843888138879249, + "learning_rate": 2.7338626164998278e-05, + "loss": 0.2868, + "num_tokens": 854389963.0, + "step": 1635 + }, + { + "epoch": 1.5272473218444342, + "grad_norm": 0.2509694150649848, + "learning_rate": 2.7252329996548153e-05, + "loss": 0.2818, + "num_tokens": 857011403.0, + "step": 1640 + }, + { + "epoch": 1.5319049836981835, + "grad_norm": 0.33386840811737556, + "learning_rate": 2.7166033828098038e-05, + "loss": 0.2837, + "num_tokens": 859632843.0, + "step": 1645 + }, + { + "epoch": 1.5365626455519328, + "grad_norm": 0.2519060001119771, + "learning_rate": 2.7079737659647913e-05, + "loss": 0.2688, + "num_tokens": 862254283.0, + "step": 1650 + }, + { + "epoch": 1.5412203074056823, + "grad_norm": 0.26036183172231986, + "learning_rate": 2.699344149119779e-05, + "loss": 0.2694, + "num_tokens": 864875723.0, + "step": 1655 + }, + { + "epoch": 1.5458779692594318, + "grad_norm": 0.39552084812808797, + "learning_rate": 2.6907145322747673e-05, + "loss": 0.273, + "num_tokens": 867482835.0, + "step": 1660 + }, + { + "epoch": 1.5505356311131813, + "grad_norm": 0.3280396515519808, + "learning_rate": 2.682084915429755e-05, + "loss": 0.2815, + "num_tokens": 870104275.0, + "step": 1665 + }, + { + "epoch": 1.5551932929669308, + "grad_norm": 0.24102481952881588, + "learning_rate": 2.673455298584743e-05, + "loss": 0.2819, + "num_tokens": 872694928.0, + "step": 1670 + }, + { + "epoch": 1.55985095482068, + "grad_norm": 0.2762279871868237, + "learning_rate": 2.664825681739731e-05, + "loss": 0.2774, + "num_tokens": 875316368.0, + "step": 1675 + }, + { + "epoch": 1.5645086166744293, + "grad_norm": 0.3001583042906396, + "learning_rate": 2.6561960648947186e-05, + "loss": 0.283, + "num_tokens": 877937808.0, + "step": 1680 + }, + { + "epoch": 1.5691662785281788, + "grad_norm": 0.275315893161812, + "learning_rate": 2.647566448049707e-05, + "loss": 0.2696, + "num_tokens": 880553728.0, + "step": 1685 + }, + { + "epoch": 1.5738239403819283, + "grad_norm": 0.28562133915152954, + "learning_rate": 2.6389368312046945e-05, + "loss": 0.2867, + "num_tokens": 883175168.0, + "step": 1690 + }, + { + "epoch": 1.5784816022356778, + "grad_norm": 0.2801444380047681, + "learning_rate": 2.6303072143596824e-05, + "loss": 0.278, + "num_tokens": 885796608.0, + "step": 1695 + }, + { + "epoch": 1.583139264089427, + "grad_norm": 0.2776860382331768, + "learning_rate": 2.6216775975146705e-05, + "loss": 0.2777, + "num_tokens": 888410029.0, + "step": 1700 + }, + { + "epoch": 1.5877969259431766, + "grad_norm": 0.2508178609607129, + "learning_rate": 2.6130479806696584e-05, + "loss": 0.2778, + "num_tokens": 891031469.0, + "step": 1705 + }, + { + "epoch": 1.5924545877969258, + "grad_norm": 0.24079007534321614, + "learning_rate": 2.6044183638246462e-05, + "loss": 0.2829, + "num_tokens": 893652909.0, + "step": 1710 + }, + { + "epoch": 1.5971122496506753, + "grad_norm": 0.24438421533644653, + "learning_rate": 2.5957887469796343e-05, + "loss": 0.2769, + "num_tokens": 896242415.0, + "step": 1715 + }, + { + "epoch": 1.6017699115044248, + "grad_norm": 0.22091161839105622, + "learning_rate": 2.587159130134622e-05, + "loss": 0.2657, + "num_tokens": 898863855.0, + "step": 1720 + }, + { + "epoch": 1.6064275733581743, + "grad_norm": 0.2611685768405385, + "learning_rate": 2.5785295132896096e-05, + "loss": 0.2823, + "num_tokens": 901485295.0, + "step": 1725 + }, + { + "epoch": 1.6110852352119236, + "grad_norm": 0.31023224431225427, + "learning_rate": 2.569899896444598e-05, + "loss": 0.2955, + "num_tokens": 904106735.0, + "step": 1730 + }, + { + "epoch": 1.6157428970656729, + "grad_norm": 0.24266065185424698, + "learning_rate": 2.5612702795995856e-05, + "loss": 0.2892, + "num_tokens": 906728175.0, + "step": 1735 + }, + { + "epoch": 1.6204005589194224, + "grad_norm": 0.22481340786142293, + "learning_rate": 2.552640662754574e-05, + "loss": 0.2724, + "num_tokens": 909349615.0, + "step": 1740 + }, + { + "epoch": 1.6250582207731719, + "grad_norm": 0.24602537155724646, + "learning_rate": 2.5440110459095616e-05, + "loss": 0.3032, + "num_tokens": 911971055.0, + "step": 1745 + }, + { + "epoch": 1.6297158826269214, + "grad_norm": 0.25533740298732005, + "learning_rate": 2.5353814290645494e-05, + "loss": 0.2896, + "num_tokens": 914580843.0, + "step": 1750 + }, + { + "epoch": 1.6343735444806708, + "grad_norm": 0.2520278922175136, + "learning_rate": 2.5267518122195376e-05, + "loss": 0.2775, + "num_tokens": 917202283.0, + "step": 1755 + }, + { + "epoch": 1.6390312063344201, + "grad_norm": 0.6074818493123854, + "learning_rate": 2.5181221953745254e-05, + "loss": 0.2739, + "num_tokens": 919823723.0, + "step": 1760 + }, + { + "epoch": 1.6436888681881694, + "grad_norm": 0.23875498925866181, + "learning_rate": 2.5094925785295132e-05, + "loss": 0.2622, + "num_tokens": 922445163.0, + "step": 1765 + }, + { + "epoch": 1.648346530041919, + "grad_norm": 0.25996583848456867, + "learning_rate": 2.5008629616845014e-05, + "loss": 0.2766, + "num_tokens": 925060687.0, + "step": 1770 + }, + { + "epoch": 1.6530041918956684, + "grad_norm": 0.6741356509956318, + "learning_rate": 2.4922333448394892e-05, + "loss": 0.2904, + "num_tokens": 927682127.0, + "step": 1775 + }, + { + "epoch": 1.6576618537494179, + "grad_norm": 0.2649759220570625, + "learning_rate": 2.483603727994477e-05, + "loss": 0.2706, + "num_tokens": 930303567.0, + "step": 1780 + }, + { + "epoch": 1.6623195156031674, + "grad_norm": 0.24250480686172254, + "learning_rate": 2.4749741111494652e-05, + "loss": 0.2808, + "num_tokens": 932925007.0, + "step": 1785 + }, + { + "epoch": 1.6669771774569166, + "grad_norm": 0.2641363849028102, + "learning_rate": 2.466344494304453e-05, + "loss": 0.2826, + "num_tokens": 935546447.0, + "step": 1790 + }, + { + "epoch": 1.671634839310666, + "grad_norm": 0.25909638449622047, + "learning_rate": 2.457714877459441e-05, + "loss": 0.2705, + "num_tokens": 938167887.0, + "step": 1795 + }, + { + "epoch": 1.6762925011644154, + "grad_norm": 0.22021948108349393, + "learning_rate": 2.4490852606144287e-05, + "loss": 0.2741, + "num_tokens": 940789327.0, + "step": 1800 + }, + { + "epoch": 1.680950163018165, + "grad_norm": 0.25132553299495947, + "learning_rate": 2.440455643769417e-05, + "loss": 0.2881, + "num_tokens": 943396878.0, + "step": 1805 + }, + { + "epoch": 1.6856078248719144, + "grad_norm": 0.25101345532688984, + "learning_rate": 2.4318260269244047e-05, + "loss": 0.2785, + "num_tokens": 946011477.0, + "step": 1810 + }, + { + "epoch": 1.6902654867256637, + "grad_norm": 0.2759051677371723, + "learning_rate": 2.4231964100793925e-05, + "loss": 0.2754, + "num_tokens": 948632917.0, + "step": 1815 + }, + { + "epoch": 1.6949231485794132, + "grad_norm": 0.23143291762750065, + "learning_rate": 2.4145667932343803e-05, + "loss": 0.2663, + "num_tokens": 951254357.0, + "step": 1820 + }, + { + "epoch": 1.6995808104331624, + "grad_norm": 0.25363871829726614, + "learning_rate": 2.4059371763893685e-05, + "loss": 0.268, + "num_tokens": 953875797.0, + "step": 1825 + }, + { + "epoch": 1.704238472286912, + "grad_norm": 0.23655353808018872, + "learning_rate": 2.3973075595443563e-05, + "loss": 0.2751, + "num_tokens": 956497237.0, + "step": 1830 + }, + { + "epoch": 1.7088961341406614, + "grad_norm": 0.23922139243346943, + "learning_rate": 2.388677942699344e-05, + "loss": 0.2817, + "num_tokens": 959085993.0, + "step": 1835 + }, + { + "epoch": 1.713553795994411, + "grad_norm": 0.2602832471502447, + "learning_rate": 2.3800483258543323e-05, + "loss": 0.2713, + "num_tokens": 961707433.0, + "step": 1840 + }, + { + "epoch": 1.7182114578481602, + "grad_norm": 0.25485339870064433, + "learning_rate": 2.37141870900932e-05, + "loss": 0.2786, + "num_tokens": 964320963.0, + "step": 1845 + }, + { + "epoch": 1.7228691197019095, + "grad_norm": 0.2899423992341778, + "learning_rate": 2.3627890921643083e-05, + "loss": 0.2919, + "num_tokens": 966942403.0, + "step": 1850 + }, + { + "epoch": 1.727526781555659, + "grad_norm": 0.2745650217168605, + "learning_rate": 2.3541594753192957e-05, + "loss": 0.2788, + "num_tokens": 969563843.0, + "step": 1855 + }, + { + "epoch": 1.7321844434094085, + "grad_norm": 0.329326878812573, + "learning_rate": 2.345529858474284e-05, + "loss": 0.2909, + "num_tokens": 972185283.0, + "step": 1860 + }, + { + "epoch": 1.736842105263158, + "grad_norm": 0.2521861694240938, + "learning_rate": 2.3369002416292717e-05, + "loss": 0.2684, + "num_tokens": 974806723.0, + "step": 1865 + }, + { + "epoch": 1.7414997671169075, + "grad_norm": 0.21998196672449688, + "learning_rate": 2.32827062478426e-05, + "loss": 0.2706, + "num_tokens": 977428163.0, + "step": 1870 + }, + { + "epoch": 1.7461574289706567, + "grad_norm": 0.2622969918570912, + "learning_rate": 2.3196410079392474e-05, + "loss": 0.2879, + "num_tokens": 980049603.0, + "step": 1875 + }, + { + "epoch": 1.750815090824406, + "grad_norm": 0.3115008184293068, + "learning_rate": 2.3110113910942355e-05, + "loss": 0.2854, + "num_tokens": 982671043.0, + "step": 1880 + }, + { + "epoch": 1.7554727526781555, + "grad_norm": 0.2868213227438428, + "learning_rate": 2.3023817742492234e-05, + "loss": 0.2814, + "num_tokens": 985291047.0, + "step": 1885 + }, + { + "epoch": 1.760130414531905, + "grad_norm": 0.2495004959502272, + "learning_rate": 2.2937521574042115e-05, + "loss": 0.2837, + "num_tokens": 987883669.0, + "step": 1890 + }, + { + "epoch": 1.7647880763856545, + "grad_norm": 0.24810813168345067, + "learning_rate": 2.285122540559199e-05, + "loss": 0.2885, + "num_tokens": 990481951.0, + "step": 1895 + }, + { + "epoch": 1.7694457382394038, + "grad_norm": 0.25530644222386045, + "learning_rate": 2.276492923714187e-05, + "loss": 0.2792, + "num_tokens": 993103391.0, + "step": 1900 + }, + { + "epoch": 1.7741034000931533, + "grad_norm": 0.23819977270988402, + "learning_rate": 2.267863306869175e-05, + "loss": 0.2844, + "num_tokens": 995694102.0, + "step": 1905 + }, + { + "epoch": 1.7787610619469025, + "grad_norm": 0.2266657031311718, + "learning_rate": 2.259233690024163e-05, + "loss": 0.2674, + "num_tokens": 998315542.0, + "step": 1910 + }, + { + "epoch": 1.783418723800652, + "grad_norm": 0.22723406455775144, + "learning_rate": 2.250604073179151e-05, + "loss": 0.2738, + "num_tokens": 1000936982.0, + "step": 1915 + }, + { + "epoch": 1.7880763856544015, + "grad_norm": 0.2603443049602763, + "learning_rate": 2.2419744563341388e-05, + "loss": 0.2775, + "num_tokens": 1003558422.0, + "step": 1920 + }, + { + "epoch": 1.792734047508151, + "grad_norm": 0.29402024633598206, + "learning_rate": 2.233344839489127e-05, + "loss": 0.2885, + "num_tokens": 1006149327.0, + "step": 1925 + }, + { + "epoch": 1.7973917093619003, + "grad_norm": 0.2440028336911219, + "learning_rate": 2.2247152226441148e-05, + "loss": 0.2675, + "num_tokens": 1008761876.0, + "step": 1930 + }, + { + "epoch": 1.8020493712156498, + "grad_norm": 0.23990807035978515, + "learning_rate": 2.2160856057991026e-05, + "loss": 0.2697, + "num_tokens": 1011383316.0, + "step": 1935 + }, + { + "epoch": 1.806707033069399, + "grad_norm": 0.22574582637141183, + "learning_rate": 2.2074559889540904e-05, + "loss": 0.276, + "num_tokens": 1014004756.0, + "step": 1940 + }, + { + "epoch": 1.8113646949231486, + "grad_norm": 0.2806755751951861, + "learning_rate": 2.1988263721090786e-05, + "loss": 0.2801, + "num_tokens": 1016604716.0, + "step": 1945 + }, + { + "epoch": 1.816022356776898, + "grad_norm": 0.22478515202508018, + "learning_rate": 2.1901967552640664e-05, + "loss": 0.2678, + "num_tokens": 1019224670.0, + "step": 1950 + }, + { + "epoch": 1.8206800186306475, + "grad_norm": 0.23339158485383402, + "learning_rate": 2.1815671384190542e-05, + "loss": 0.2739, + "num_tokens": 1021846110.0, + "step": 1955 + }, + { + "epoch": 1.8253376804843968, + "grad_norm": 0.2447423723444364, + "learning_rate": 2.172937521574042e-05, + "loss": 0.2777, + "num_tokens": 1024467550.0, + "step": 1960 + }, + { + "epoch": 1.829995342338146, + "grad_norm": 0.322101332959102, + "learning_rate": 2.1643079047290302e-05, + "loss": 0.2743, + "num_tokens": 1027061863.0, + "step": 1965 + }, + { + "epoch": 1.8346530041918956, + "grad_norm": 0.2950678541012179, + "learning_rate": 2.155678287884018e-05, + "loss": 0.2643, + "num_tokens": 1029683303.0, + "step": 1970 + }, + { + "epoch": 1.839310666045645, + "grad_norm": 0.27257343724970284, + "learning_rate": 2.147048671039006e-05, + "loss": 0.2813, + "num_tokens": 1032304743.0, + "step": 1975 + }, + { + "epoch": 1.8439683278993946, + "grad_norm": 0.23708782005593013, + "learning_rate": 2.1384190541939937e-05, + "loss": 0.2696, + "num_tokens": 1034926183.0, + "step": 1980 + }, + { + "epoch": 1.848625989753144, + "grad_norm": 0.23810036151619607, + "learning_rate": 2.129789437348982e-05, + "loss": 0.2752, + "num_tokens": 1037538252.0, + "step": 1985 + }, + { + "epoch": 1.8532836516068933, + "grad_norm": 0.22131956602133956, + "learning_rate": 2.1211598205039697e-05, + "loss": 0.2816, + "num_tokens": 1040159692.0, + "step": 1990 + }, + { + "epoch": 1.8579413134606426, + "grad_norm": 0.26221654380992987, + "learning_rate": 2.1125302036589575e-05, + "loss": 0.276, + "num_tokens": 1042781132.0, + "step": 1995 + }, + { + "epoch": 1.8625989753143921, + "grad_norm": 0.2763721434630872, + "learning_rate": 2.1039005868139457e-05, + "loss": 0.2743, + "num_tokens": 1045402572.0, + "step": 2000 + }, + { + "epoch": 1.8672566371681416, + "grad_norm": 0.2782332025416994, + "learning_rate": 2.0952709699689335e-05, + "loss": 0.2823, + "num_tokens": 1048018446.0, + "step": 2005 + }, + { + "epoch": 1.871914299021891, + "grad_norm": 0.23803614005307103, + "learning_rate": 2.0866413531239216e-05, + "loss": 0.2686, + "num_tokens": 1050639886.0, + "step": 2010 + }, + { + "epoch": 1.8765719608756404, + "grad_norm": 0.22674206087841164, + "learning_rate": 2.078011736278909e-05, + "loss": 0.2746, + "num_tokens": 1053261326.0, + "step": 2015 + }, + { + "epoch": 1.8812296227293899, + "grad_norm": 0.22725091660452684, + "learning_rate": 2.0693821194338973e-05, + "loss": 0.2735, + "num_tokens": 1055871697.0, + "step": 2020 + }, + { + "epoch": 1.8858872845831391, + "grad_norm": 0.24433878595752298, + "learning_rate": 2.060752502588885e-05, + "loss": 0.2815, + "num_tokens": 1058493137.0, + "step": 2025 + }, + { + "epoch": 1.8905449464368886, + "grad_norm": 0.2666883415037318, + "learning_rate": 2.0521228857438733e-05, + "loss": 0.2753, + "num_tokens": 1061114577.0, + "step": 2030 + }, + { + "epoch": 1.8952026082906381, + "grad_norm": 0.2569040671161156, + "learning_rate": 2.0434932688988608e-05, + "loss": 0.2721, + "num_tokens": 1063736017.0, + "step": 2035 + }, + { + "epoch": 1.8998602701443876, + "grad_norm": 0.25149835872795157, + "learning_rate": 2.034863652053849e-05, + "loss": 0.2787, + "num_tokens": 1066357457.0, + "step": 2040 + }, + { + "epoch": 1.904517931998137, + "grad_norm": 0.2361648890464829, + "learning_rate": 2.0262340352088367e-05, + "loss": 0.277, + "num_tokens": 1068978897.0, + "step": 2045 + }, + { + "epoch": 1.9091755938518864, + "grad_norm": 0.25758614769547705, + "learning_rate": 2.017604418363825e-05, + "loss": 0.2832, + "num_tokens": 1071600337.0, + "step": 2050 + }, + { + "epoch": 1.9138332557056357, + "grad_norm": 0.23652596339114235, + "learning_rate": 2.0089748015188127e-05, + "loss": 0.2886, + "num_tokens": 1074208623.0, + "step": 2055 + }, + { + "epoch": 1.9184909175593852, + "grad_norm": 0.2578336365580039, + "learning_rate": 2.0003451846738005e-05, + "loss": 0.2751, + "num_tokens": 1076785605.0, + "step": 2060 + }, + { + "epoch": 1.9231485794131347, + "grad_norm": 0.23243843826265423, + "learning_rate": 1.9917155678287887e-05, + "loss": 0.2645, + "num_tokens": 1079407045.0, + "step": 2065 + }, + { + "epoch": 1.9278062412668842, + "grad_norm": 0.25865227066679375, + "learning_rate": 1.9830859509837765e-05, + "loss": 0.2692, + "num_tokens": 1082028485.0, + "step": 2070 + }, + { + "epoch": 1.9324639031206334, + "grad_norm": 0.2347806044881937, + "learning_rate": 1.9744563341387643e-05, + "loss": 0.2747, + "num_tokens": 1084649925.0, + "step": 2075 + }, + { + "epoch": 1.9371215649743827, + "grad_norm": 0.27455209060598385, + "learning_rate": 1.9658267172937522e-05, + "loss": 0.2751, + "num_tokens": 1087271365.0, + "step": 2080 + }, + { + "epoch": 1.9417792268281322, + "grad_norm": 0.22749532966422994, + "learning_rate": 1.9571971004487403e-05, + "loss": 0.2833, + "num_tokens": 1089860165.0, + "step": 2085 + }, + { + "epoch": 1.9464368886818817, + "grad_norm": 0.25182841487010565, + "learning_rate": 1.948567483603728e-05, + "loss": 0.2674, + "num_tokens": 1092481605.0, + "step": 2090 + }, + { + "epoch": 1.9510945505356312, + "grad_norm": 0.2315781981216672, + "learning_rate": 1.939937866758716e-05, + "loss": 0.2698, + "num_tokens": 1095082526.0, + "step": 2095 + }, + { + "epoch": 1.9557522123893807, + "grad_norm": 0.23764701466290206, + "learning_rate": 1.9313082499137038e-05, + "loss": 0.2771, + "num_tokens": 1097703966.0, + "step": 2100 + }, + { + "epoch": 1.96040987424313, + "grad_norm": 0.24584826617890893, + "learning_rate": 1.922678633068692e-05, + "loss": 0.286, + "num_tokens": 1100325406.0, + "step": 2105 + }, + { + "epoch": 1.9650675360968792, + "grad_norm": 0.22948941314974505, + "learning_rate": 1.9140490162236798e-05, + "loss": 0.2714, + "num_tokens": 1102946846.0, + "step": 2110 + }, + { + "epoch": 1.9697251979506287, + "grad_norm": 0.24532426266235913, + "learning_rate": 1.9054193993786676e-05, + "loss": 0.2742, + "num_tokens": 1105554920.0, + "step": 2115 + }, + { + "epoch": 1.9743828598043782, + "grad_norm": 0.23211658563196494, + "learning_rate": 1.8967897825336554e-05, + "loss": 0.275, + "num_tokens": 1108176360.0, + "step": 2120 + }, + { + "epoch": 1.9790405216581277, + "grad_norm": 0.22106706200612478, + "learning_rate": 1.8881601656886436e-05, + "loss": 0.2783, + "num_tokens": 1110768880.0, + "step": 2125 + }, + { + "epoch": 1.983698183511877, + "grad_norm": 0.24816638342159197, + "learning_rate": 1.8795305488436314e-05, + "loss": 0.2769, + "num_tokens": 1113390320.0, + "step": 2130 + }, + { + "epoch": 1.9883558453656265, + "grad_norm": 0.26279439526160714, + "learning_rate": 1.8709009319986192e-05, + "loss": 0.279, + "num_tokens": 1116011760.0, + "step": 2135 + }, + { + "epoch": 1.9930135072193758, + "grad_norm": 0.22476550281934327, + "learning_rate": 1.8622713151536074e-05, + "loss": 0.2755, + "num_tokens": 1118633200.0, + "step": 2140 + }, + { + "epoch": 1.9976711690731253, + "grad_norm": 0.22321618125402085, + "learning_rate": 1.8536416983085952e-05, + "loss": 0.2841, + "num_tokens": 1121254640.0, + "step": 2145 + }, + { + "epoch": 2.0018630647415, + "grad_norm": 0.22811789448737538, + "learning_rate": 1.8450120814635834e-05, + "loss": 0.2459, + "num_tokens": 1123444248.0, + "step": 2150 + }, + { + "epoch": 2.0065207265952494, + "grad_norm": 0.23763390471988635, + "learning_rate": 1.836382464618571e-05, + "loss": 0.2109, + "num_tokens": 1126056407.0, + "step": 2155 + }, + { + "epoch": 2.0111783884489984, + "grad_norm": 0.2159028160161467, + "learning_rate": 1.827752847773559e-05, + "loss": 0.2033, + "num_tokens": 1128677847.0, + "step": 2160 + }, + { + "epoch": 2.015836050302748, + "grad_norm": 0.2354086898913808, + "learning_rate": 1.819123230928547e-05, + "loss": 0.1927, + "num_tokens": 1131289728.0, + "step": 2165 + }, + { + "epoch": 2.0204937121564974, + "grad_norm": 0.22993245593936573, + "learning_rate": 1.810493614083535e-05, + "loss": 0.1869, + "num_tokens": 1133911168.0, + "step": 2170 + }, + { + "epoch": 2.025151374010247, + "grad_norm": 0.2788929387492099, + "learning_rate": 1.8018639972385225e-05, + "loss": 0.1958, + "num_tokens": 1136511128.0, + "step": 2175 + }, + { + "epoch": 2.0298090358639964, + "grad_norm": 0.28967455780722556, + "learning_rate": 1.7932343803935107e-05, + "loss": 0.1805, + "num_tokens": 1139132568.0, + "step": 2180 + }, + { + "epoch": 2.034466697717746, + "grad_norm": 0.24813904774828296, + "learning_rate": 1.7846047635484985e-05, + "loss": 0.1903, + "num_tokens": 1141754008.0, + "step": 2185 + }, + { + "epoch": 2.039124359571495, + "grad_norm": 0.26291754418443875, + "learning_rate": 1.7759751467034866e-05, + "loss": 0.2017, + "num_tokens": 1144364017.0, + "step": 2190 + }, + { + "epoch": 2.0437820214252445, + "grad_norm": 0.21454863669215662, + "learning_rate": 1.767345529858474e-05, + "loss": 0.1843, + "num_tokens": 1146985457.0, + "step": 2195 + }, + { + "epoch": 2.048439683278994, + "grad_norm": 0.21582532066367777, + "learning_rate": 1.7587159130134623e-05, + "loss": 0.1804, + "num_tokens": 1149576149.0, + "step": 2200 + }, + { + "epoch": 2.0530973451327434, + "grad_norm": 0.21932933980572245, + "learning_rate": 1.75008629616845e-05, + "loss": 0.1963, + "num_tokens": 1152197589.0, + "step": 2205 + }, + { + "epoch": 2.057755006986493, + "grad_norm": 0.2151303685730624, + "learning_rate": 1.7414566793234383e-05, + "loss": 0.1871, + "num_tokens": 1154799708.0, + "step": 2210 + }, + { + "epoch": 2.062412668840242, + "grad_norm": 0.2222160470869714, + "learning_rate": 1.732827062478426e-05, + "loss": 0.1991, + "num_tokens": 1157421148.0, + "step": 2215 + }, + { + "epoch": 2.0670703306939915, + "grad_norm": 0.3443027048047173, + "learning_rate": 1.724197445633414e-05, + "loss": 0.1861, + "num_tokens": 1160042588.0, + "step": 2220 + }, + { + "epoch": 2.071727992547741, + "grad_norm": 0.26261630346021714, + "learning_rate": 1.715567828788402e-05, + "loss": 0.1853, + "num_tokens": 1162664028.0, + "step": 2225 + }, + { + "epoch": 2.0763856544014905, + "grad_norm": 0.19977835774642955, + "learning_rate": 1.70693821194339e-05, + "loss": 0.1959, + "num_tokens": 1165285468.0, + "step": 2230 + }, + { + "epoch": 2.08104331625524, + "grad_norm": 0.21770826209782151, + "learning_rate": 1.6983085950983777e-05, + "loss": 0.1851, + "num_tokens": 1167906908.0, + "step": 2235 + }, + { + "epoch": 2.0857009781089895, + "grad_norm": 0.208891787309407, + "learning_rate": 1.6896789782533655e-05, + "loss": 0.1925, + "num_tokens": 1170528348.0, + "step": 2240 + }, + { + "epoch": 2.0903586399627385, + "grad_norm": 0.2131965536139105, + "learning_rate": 1.6810493614083537e-05, + "loss": 0.1899, + "num_tokens": 1173119003.0, + "step": 2245 + }, + { + "epoch": 2.095016301816488, + "grad_norm": 0.2036260100247772, + "learning_rate": 1.6724197445633415e-05, + "loss": 0.1785, + "num_tokens": 1175740443.0, + "step": 2250 + }, + { + "epoch": 2.0996739636702375, + "grad_norm": 0.29436739494405906, + "learning_rate": 1.6637901277183294e-05, + "loss": 0.1959, + "num_tokens": 1178339533.0, + "step": 2255 + }, + { + "epoch": 2.104331625523987, + "grad_norm": 0.22398528196888434, + "learning_rate": 1.6551605108733172e-05, + "loss": 0.1844, + "num_tokens": 1180960973.0, + "step": 2260 + }, + { + "epoch": 2.1089892873777365, + "grad_norm": 0.4929934728417884, + "learning_rate": 1.6465308940283053e-05, + "loss": 0.2028, + "num_tokens": 1183560581.0, + "step": 2265 + }, + { + "epoch": 2.113646949231486, + "grad_norm": 0.23022866493235755, + "learning_rate": 1.637901277183293e-05, + "loss": 0.1898, + "num_tokens": 1186182021.0, + "step": 2270 + }, + { + "epoch": 2.118304611085235, + "grad_norm": 0.22441316024216518, + "learning_rate": 1.629271660338281e-05, + "loss": 0.188, + "num_tokens": 1188803461.0, + "step": 2275 + }, + { + "epoch": 2.1229622729389845, + "grad_norm": 0.23904511220914146, + "learning_rate": 1.620642043493269e-05, + "loss": 0.1869, + "num_tokens": 1191413832.0, + "step": 2280 + }, + { + "epoch": 2.127619934792734, + "grad_norm": 0.20446264923991703, + "learning_rate": 1.612012426648257e-05, + "loss": 0.1872, + "num_tokens": 1194013826.0, + "step": 2285 + }, + { + "epoch": 2.1322775966464835, + "grad_norm": 0.23153824701671225, + "learning_rate": 1.6033828098032448e-05, + "loss": 0.1946, + "num_tokens": 1196635266.0, + "step": 2290 + }, + { + "epoch": 2.136935258500233, + "grad_norm": 0.2227822286301486, + "learning_rate": 1.5947531929582326e-05, + "loss": 0.1908, + "num_tokens": 1199233871.0, + "step": 2295 + }, + { + "epoch": 2.1415929203539825, + "grad_norm": 0.2482664314610804, + "learning_rate": 1.5861235761132208e-05, + "loss": 0.1982, + "num_tokens": 1201855311.0, + "step": 2300 + }, + { + "epoch": 2.1462505822077316, + "grad_norm": 0.21568483935394447, + "learning_rate": 1.5774939592682086e-05, + "loss": 0.1948, + "num_tokens": 1204475607.0, + "step": 2305 + }, + { + "epoch": 2.150908244061481, + "grad_norm": 0.20282058165536931, + "learning_rate": 1.5688643424231964e-05, + "loss": 0.1828, + "num_tokens": 1207097047.0, + "step": 2310 + }, + { + "epoch": 2.1555659059152306, + "grad_norm": 0.6030692140488413, + "learning_rate": 1.5602347255781842e-05, + "loss": 0.1926, + "num_tokens": 1209685847.0, + "step": 2315 + }, + { + "epoch": 2.16022356776898, + "grad_norm": 0.2045510595258148, + "learning_rate": 1.5516051087331724e-05, + "loss": 0.1852, + "num_tokens": 1212307287.0, + "step": 2320 + }, + { + "epoch": 2.1648812296227296, + "grad_norm": 0.2229396176932403, + "learning_rate": 1.5429754918881602e-05, + "loss": 0.1929, + "num_tokens": 1214920560.0, + "step": 2325 + }, + { + "epoch": 2.1695388914764786, + "grad_norm": 0.22741654294869426, + "learning_rate": 1.534345875043148e-05, + "loss": 0.1841, + "num_tokens": 1217520276.0, + "step": 2330 + }, + { + "epoch": 2.174196553330228, + "grad_norm": 0.2218804396657255, + "learning_rate": 1.525716258198136e-05, + "loss": 0.1978, + "num_tokens": 1220141716.0, + "step": 2335 + }, + { + "epoch": 2.1788542151839776, + "grad_norm": 0.22397360159092755, + "learning_rate": 1.517086641353124e-05, + "loss": 0.1877, + "num_tokens": 1222742637.0, + "step": 2340 + }, + { + "epoch": 2.183511877037727, + "grad_norm": 0.22903346357382584, + "learning_rate": 1.508457024508112e-05, + "loss": 0.1926, + "num_tokens": 1225364077.0, + "step": 2345 + }, + { + "epoch": 2.1881695388914766, + "grad_norm": 0.22681455304615486, + "learning_rate": 1.4998274076630997e-05, + "loss": 0.1927, + "num_tokens": 1227985517.0, + "step": 2350 + }, + { + "epoch": 2.192827200745226, + "grad_norm": 0.21157141925014325, + "learning_rate": 1.4911977908180877e-05, + "loss": 0.1851, + "num_tokens": 1230593185.0, + "step": 2355 + }, + { + "epoch": 2.197484862598975, + "grad_norm": 0.24916960595742724, + "learning_rate": 1.4825681739730757e-05, + "loss": 0.1935, + "num_tokens": 1233214625.0, + "step": 2360 + }, + { + "epoch": 2.2021425244527246, + "grad_norm": 0.25376290664329243, + "learning_rate": 1.4739385571280637e-05, + "loss": 0.2031, + "num_tokens": 1235780896.0, + "step": 2365 + }, + { + "epoch": 2.206800186306474, + "grad_norm": 0.20737450540956834, + "learning_rate": 1.4653089402830513e-05, + "loss": 0.1909, + "num_tokens": 1238372031.0, + "step": 2370 + }, + { + "epoch": 2.2114578481602236, + "grad_norm": 0.22617271594955698, + "learning_rate": 1.4566793234380393e-05, + "loss": 0.1863, + "num_tokens": 1240993471.0, + "step": 2375 + }, + { + "epoch": 2.216115510013973, + "grad_norm": 0.27615735313408574, + "learning_rate": 1.4480497065930273e-05, + "loss": 0.1945, + "num_tokens": 1243606026.0, + "step": 2380 + }, + { + "epoch": 2.2207731718677226, + "grad_norm": 0.2859249546685766, + "learning_rate": 1.4394200897480153e-05, + "loss": 0.2029, + "num_tokens": 1246227466.0, + "step": 2385 + }, + { + "epoch": 2.2254308337214717, + "grad_norm": 0.19939388937682523, + "learning_rate": 1.4307904729030031e-05, + "loss": 0.2021, + "num_tokens": 1248828187.0, + "step": 2390 + }, + { + "epoch": 2.230088495575221, + "grad_norm": 0.21160041897399634, + "learning_rate": 1.4221608560579911e-05, + "loss": 0.1868, + "num_tokens": 1251449627.0, + "step": 2395 + }, + { + "epoch": 2.2347461574289706, + "grad_norm": 0.22235345505947743, + "learning_rate": 1.4135312392129791e-05, + "loss": 0.1839, + "num_tokens": 1254056658.0, + "step": 2400 + }, + { + "epoch": 2.23940381928272, + "grad_norm": 0.23712117694873538, + "learning_rate": 1.404901622367967e-05, + "loss": 0.1911, + "num_tokens": 1256678098.0, + "step": 2405 + }, + { + "epoch": 2.2440614811364696, + "grad_norm": 0.24561430494734446, + "learning_rate": 1.3962720055229547e-05, + "loss": 0.1897, + "num_tokens": 1259299538.0, + "step": 2410 + }, + { + "epoch": 2.248719142990219, + "grad_norm": 0.20679953059186723, + "learning_rate": 1.3876423886779427e-05, + "loss": 0.1905, + "num_tokens": 1261907824.0, + "step": 2415 + }, + { + "epoch": 2.253376804843968, + "grad_norm": 0.22679903703760337, + "learning_rate": 1.3790127718329307e-05, + "loss": 0.1904, + "num_tokens": 1264522423.0, + "step": 2420 + }, + { + "epoch": 2.2580344666977177, + "grad_norm": 0.21852291008494948, + "learning_rate": 1.3703831549879187e-05, + "loss": 0.184, + "num_tokens": 1267143863.0, + "step": 2425 + }, + { + "epoch": 2.262692128551467, + "grad_norm": 0.20205416517381505, + "learning_rate": 1.3617535381429064e-05, + "loss": 0.1775, + "num_tokens": 1269765303.0, + "step": 2430 + }, + { + "epoch": 2.2673497904052167, + "grad_norm": 0.21402822625968712, + "learning_rate": 1.3531239212978944e-05, + "loss": 0.1926, + "num_tokens": 1272386743.0, + "step": 2435 + }, + { + "epoch": 2.272007452258966, + "grad_norm": 0.21251576960373053, + "learning_rate": 1.3444943044528824e-05, + "loss": 0.1894, + "num_tokens": 1274982459.0, + "step": 2440 + }, + { + "epoch": 2.276665114112715, + "grad_norm": 0.19438570118120085, + "learning_rate": 1.3358646876078703e-05, + "loss": 0.1864, + "num_tokens": 1277585298.0, + "step": 2445 + }, + { + "epoch": 2.2813227759664647, + "grad_norm": 0.1969520337468591, + "learning_rate": 1.327235070762858e-05, + "loss": 0.1834, + "num_tokens": 1280205815.0, + "step": 2450 + }, + { + "epoch": 2.285980437820214, + "grad_norm": 0.19957634159837317, + "learning_rate": 1.318605453917846e-05, + "loss": 0.1919, + "num_tokens": 1282802119.0, + "step": 2455 + }, + { + "epoch": 2.2906380996739637, + "grad_norm": 0.23104800649747617, + "learning_rate": 1.309975837072834e-05, + "loss": 0.1924, + "num_tokens": 1285414071.0, + "step": 2460 + }, + { + "epoch": 2.295295761527713, + "grad_norm": 0.24111804075309679, + "learning_rate": 1.301346220227822e-05, + "loss": 0.1987, + "num_tokens": 1288023894.0, + "step": 2465 + }, + { + "epoch": 2.2999534233814627, + "grad_norm": 0.21793206443287055, + "learning_rate": 1.2927166033828098e-05, + "loss": 0.1872, + "num_tokens": 1290645334.0, + "step": 2470 + }, + { + "epoch": 2.3046110852352117, + "grad_norm": 0.2114943769815432, + "learning_rate": 1.2840869865377978e-05, + "loss": 0.1974, + "num_tokens": 1293258755.0, + "step": 2475 + }, + { + "epoch": 2.3092687470889612, + "grad_norm": 0.2091634629424366, + "learning_rate": 1.2754573696927858e-05, + "loss": 0.1832, + "num_tokens": 1295880195.0, + "step": 2480 + }, + { + "epoch": 2.3139264089427107, + "grad_norm": 0.23428714293181593, + "learning_rate": 1.2668277528477738e-05, + "loss": 0.1894, + "num_tokens": 1298501635.0, + "step": 2485 + }, + { + "epoch": 2.3185840707964602, + "grad_norm": 0.2160517380096932, + "learning_rate": 1.2581981360027614e-05, + "loss": 0.1999, + "num_tokens": 1301121639.0, + "step": 2490 + }, + { + "epoch": 2.3232417326502097, + "grad_norm": 0.2137407041544956, + "learning_rate": 1.2495685191577494e-05, + "loss": 0.1899, + "num_tokens": 1303743079.0, + "step": 2495 + }, + { + "epoch": 2.3278993945039588, + "grad_norm": 0.19309461413453677, + "learning_rate": 1.2409389023127374e-05, + "loss": 0.1802, + "num_tokens": 1306327934.0, + "step": 2500 + }, + { + "epoch": 2.3325570563577083, + "grad_norm": 0.21891242654845594, + "learning_rate": 1.2323092854677252e-05, + "loss": 0.1919, + "num_tokens": 1308949374.0, + "step": 2505 + }, + { + "epoch": 2.3372147182114578, + "grad_norm": 0.23605025676718958, + "learning_rate": 1.2236796686227132e-05, + "loss": 0.1905, + "num_tokens": 1311570814.0, + "step": 2510 + }, + { + "epoch": 2.3418723800652073, + "grad_norm": 0.21455241286434934, + "learning_rate": 1.215050051777701e-05, + "loss": 0.1932, + "num_tokens": 1314192254.0, + "step": 2515 + }, + { + "epoch": 2.3465300419189568, + "grad_norm": 0.2320284991032439, + "learning_rate": 1.206420434932689e-05, + "loss": 0.1854, + "num_tokens": 1316784876.0, + "step": 2520 + }, + { + "epoch": 2.3511877037727063, + "grad_norm": 0.25293497517809127, + "learning_rate": 1.1977908180876769e-05, + "loss": 0.1918, + "num_tokens": 1319406316.0, + "step": 2525 + }, + { + "epoch": 2.3558453656264557, + "grad_norm": 0.2146459615030723, + "learning_rate": 1.1891612012426649e-05, + "loss": 0.1907, + "num_tokens": 1322027756.0, + "step": 2530 + }, + { + "epoch": 2.360503027480205, + "grad_norm": 0.21794216655250306, + "learning_rate": 1.1805315843976528e-05, + "loss": 0.1891, + "num_tokens": 1324649196.0, + "step": 2535 + }, + { + "epoch": 2.3651606893339543, + "grad_norm": 0.21526928636835807, + "learning_rate": 1.1719019675526408e-05, + "loss": 0.1816, + "num_tokens": 1327270636.0, + "step": 2540 + }, + { + "epoch": 2.369818351187704, + "grad_norm": 0.2009434671686828, + "learning_rate": 1.1632723507076287e-05, + "loss": 0.1911, + "num_tokens": 1329892076.0, + "step": 2545 + }, + { + "epoch": 2.3744760130414533, + "grad_norm": 0.21308857609787304, + "learning_rate": 1.1546427338626167e-05, + "loss": 0.1873, + "num_tokens": 1332513516.0, + "step": 2550 + }, + { + "epoch": 2.3791336748952028, + "grad_norm": 0.20723679300713657, + "learning_rate": 1.1460131170176045e-05, + "loss": 0.1869, + "num_tokens": 1335134956.0, + "step": 2555 + }, + { + "epoch": 2.383791336748952, + "grad_norm": 0.222911629634698, + "learning_rate": 1.1373835001725925e-05, + "loss": 0.1934, + "num_tokens": 1337756396.0, + "step": 2560 + }, + { + "epoch": 2.3884489986027013, + "grad_norm": 0.2147394848596357, + "learning_rate": 1.1287538833275803e-05, + "loss": 0.1881, + "num_tokens": 1340374946.0, + "step": 2565 + }, + { + "epoch": 2.393106660456451, + "grad_norm": 0.22327545683469197, + "learning_rate": 1.1201242664825683e-05, + "loss": 0.2006, + "num_tokens": 1342996386.0, + "step": 2570 + }, + { + "epoch": 2.3977643223102003, + "grad_norm": 0.2034574667765081, + "learning_rate": 1.1114946496375561e-05, + "loss": 0.1951, + "num_tokens": 1345617826.0, + "step": 2575 + }, + { + "epoch": 2.40242198416395, + "grad_norm": 0.2113918902606263, + "learning_rate": 1.1028650327925441e-05, + "loss": 0.1909, + "num_tokens": 1348239266.0, + "step": 2580 + }, + { + "epoch": 2.4070796460176993, + "grad_norm": 0.22419572087626496, + "learning_rate": 1.094235415947532e-05, + "loss": 0.1866, + "num_tokens": 1350860706.0, + "step": 2585 + }, + { + "epoch": 2.4117373078714484, + "grad_norm": 0.19006705350069172, + "learning_rate": 1.0856057991025199e-05, + "loss": 0.1763, + "num_tokens": 1353482146.0, + "step": 2590 + }, + { + "epoch": 2.416394969725198, + "grad_norm": 0.24584362708804106, + "learning_rate": 1.0769761822575077e-05, + "loss": 0.2, + "num_tokens": 1356044035.0, + "step": 2595 + }, + { + "epoch": 2.4210526315789473, + "grad_norm": 0.21775943305708015, + "learning_rate": 1.0683465654124957e-05, + "loss": 0.2008, + "num_tokens": 1358654903.0, + "step": 2600 + }, + { + "epoch": 2.425710293432697, + "grad_norm": 0.19777137481318433, + "learning_rate": 1.0597169485674835e-05, + "loss": 0.1848, + "num_tokens": 1361276343.0, + "step": 2605 + }, + { + "epoch": 2.4303679552864463, + "grad_norm": 0.20775299467322017, + "learning_rate": 1.0510873317224715e-05, + "loss": 0.1865, + "num_tokens": 1363897783.0, + "step": 2610 + }, + { + "epoch": 2.4350256171401954, + "grad_norm": 0.19894049766984748, + "learning_rate": 1.0424577148774595e-05, + "loss": 0.1888, + "num_tokens": 1366519223.0, + "step": 2615 + }, + { + "epoch": 2.439683278993945, + "grad_norm": 0.23413570582928064, + "learning_rate": 1.0338280980324475e-05, + "loss": 0.1982, + "num_tokens": 1369124851.0, + "step": 2620 + }, + { + "epoch": 2.4443409408476944, + "grad_norm": 0.21657572328477895, + "learning_rate": 1.0251984811874353e-05, + "loss": 0.1819, + "num_tokens": 1371746291.0, + "step": 2625 + }, + { + "epoch": 2.448998602701444, + "grad_norm": 0.20427064714545917, + "learning_rate": 1.0165688643424233e-05, + "loss": 0.1957, + "num_tokens": 1374367731.0, + "step": 2630 + }, + { + "epoch": 2.4536562645551934, + "grad_norm": 0.2127507944462461, + "learning_rate": 1.0079392474974112e-05, + "loss": 0.1885, + "num_tokens": 1376989171.0, + "step": 2635 + }, + { + "epoch": 2.458313926408943, + "grad_norm": 0.22496735588127278, + "learning_rate": 9.993096306523992e-06, + "loss": 0.1915, + "num_tokens": 1379610611.0, + "step": 2640 + }, + { + "epoch": 2.4629715882626924, + "grad_norm": 0.250954777611335, + "learning_rate": 9.90680013807387e-06, + "loss": 0.1965, + "num_tokens": 1382232051.0, + "step": 2645 + }, + { + "epoch": 2.4676292501164414, + "grad_norm": 0.239020428214204, + "learning_rate": 9.82050396962375e-06, + "loss": 0.1909, + "num_tokens": 1384853491.0, + "step": 2650 + }, + { + "epoch": 2.472286911970191, + "grad_norm": 0.21219394873292483, + "learning_rate": 9.734207801173628e-06, + "loss": 0.1853, + "num_tokens": 1387439705.0, + "step": 2655 + }, + { + "epoch": 2.4769445738239404, + "grad_norm": 0.2339132165376453, + "learning_rate": 9.647911632723508e-06, + "loss": 0.1859, + "num_tokens": 1390055579.0, + "step": 2660 + }, + { + "epoch": 2.48160223567769, + "grad_norm": 0.20931019992718092, + "learning_rate": 9.561615464273386e-06, + "loss": 0.1946, + "num_tokens": 1392677019.0, + "step": 2665 + }, + { + "epoch": 2.4862598975314394, + "grad_norm": 0.21664209547112198, + "learning_rate": 9.475319295823266e-06, + "loss": 0.1879, + "num_tokens": 1395298459.0, + "step": 2670 + }, + { + "epoch": 2.4909175593851884, + "grad_norm": 0.21377592639772283, + "learning_rate": 9.389023127373144e-06, + "loss": 0.1873, + "num_tokens": 1397919899.0, + "step": 2675 + }, + { + "epoch": 2.495575221238938, + "grad_norm": 0.2100788699209234, + "learning_rate": 9.302726958923024e-06, + "loss": 0.1894, + "num_tokens": 1400541339.0, + "step": 2680 + }, + { + "epoch": 2.5002328830926874, + "grad_norm": 0.24898127480817744, + "learning_rate": 9.216430790472904e-06, + "loss": 0.1921, + "num_tokens": 1403162779.0, + "step": 2685 + }, + { + "epoch": 2.504890544946437, + "grad_norm": 0.2679217103887551, + "learning_rate": 9.130134622022784e-06, + "loss": 0.1867, + "num_tokens": 1405784219.0, + "step": 2690 + }, + { + "epoch": 2.5095482068001864, + "grad_norm": 0.264365285211875, + "learning_rate": 9.043838453572662e-06, + "loss": 0.1878, + "num_tokens": 1408405659.0, + "step": 2695 + }, + { + "epoch": 2.514205868653936, + "grad_norm": 0.2177074226742054, + "learning_rate": 8.957542285122542e-06, + "loss": 0.1868, + "num_tokens": 1411027099.0, + "step": 2700 + }, + { + "epoch": 2.5188635305076854, + "grad_norm": 0.22624315481556287, + "learning_rate": 8.87124611667242e-06, + "loss": 0.1876, + "num_tokens": 1413648539.0, + "step": 2705 + }, + { + "epoch": 2.5235211923614345, + "grad_norm": 0.1978006596252754, + "learning_rate": 8.7849499482223e-06, + "loss": 0.1905, + "num_tokens": 1416269979.0, + "step": 2710 + }, + { + "epoch": 2.528178854215184, + "grad_norm": 0.274994491294196, + "learning_rate": 8.698653779772179e-06, + "loss": 0.1884, + "num_tokens": 1418891419.0, + "step": 2715 + }, + { + "epoch": 2.5328365160689335, + "grad_norm": 0.19469792288627025, + "learning_rate": 8.612357611322058e-06, + "loss": 0.1949, + "num_tokens": 1421512859.0, + "step": 2720 + }, + { + "epoch": 2.537494177922683, + "grad_norm": 0.20344558223616066, + "learning_rate": 8.526061442871937e-06, + "loss": 0.1904, + "num_tokens": 1424134299.0, + "step": 2725 + }, + { + "epoch": 2.542151839776432, + "grad_norm": 0.20543803618958464, + "learning_rate": 8.439765274421817e-06, + "loss": 0.1947, + "num_tokens": 1426729315.0, + "step": 2730 + }, + { + "epoch": 2.5468095016301815, + "grad_norm": 0.20586834460357978, + "learning_rate": 8.353469105971695e-06, + "loss": 0.1799, + "num_tokens": 1429350755.0, + "step": 2735 + }, + { + "epoch": 2.551467163483931, + "grad_norm": 0.20869595757236642, + "learning_rate": 8.267172937521575e-06, + "loss": 0.1849, + "num_tokens": 1431972195.0, + "step": 2740 + }, + { + "epoch": 2.5561248253376805, + "grad_norm": 0.20469546705647887, + "learning_rate": 8.180876769071453e-06, + "loss": 0.193, + "num_tokens": 1434593635.0, + "step": 2745 + }, + { + "epoch": 2.56078248719143, + "grad_norm": 0.19749997870794891, + "learning_rate": 8.094580600621333e-06, + "loss": 0.1906, + "num_tokens": 1437215075.0, + "step": 2750 + }, + { + "epoch": 2.5654401490451795, + "grad_norm": 0.21537997550568386, + "learning_rate": 8.008284432171211e-06, + "loss": 0.1905, + "num_tokens": 1439836515.0, + "step": 2755 + }, + { + "epoch": 2.570097810898929, + "grad_norm": 0.227592429204148, + "learning_rate": 7.921988263721091e-06, + "loss": 0.1821, + "num_tokens": 1442457955.0, + "step": 2760 + }, + { + "epoch": 2.574755472752678, + "grad_norm": 0.234230256816076, + "learning_rate": 7.835692095270971e-06, + "loss": 0.1865, + "num_tokens": 1445062211.0, + "step": 2765 + }, + { + "epoch": 2.5794131346064275, + "grad_norm": 0.25861680354818406, + "learning_rate": 7.749395926820851e-06, + "loss": 0.1848, + "num_tokens": 1447683651.0, + "step": 2770 + }, + { + "epoch": 2.584070796460177, + "grad_norm": 0.23014158285970068, + "learning_rate": 7.663099758370729e-06, + "loss": 0.1826, + "num_tokens": 1450303010.0, + "step": 2775 + }, + { + "epoch": 2.5887284583139265, + "grad_norm": 0.20030501496288292, + "learning_rate": 7.576803589920608e-06, + "loss": 0.1937, + "num_tokens": 1452924450.0, + "step": 2780 + }, + { + "epoch": 2.5933861201676756, + "grad_norm": 0.21189324719032693, + "learning_rate": 7.490507421470487e-06, + "loss": 0.1879, + "num_tokens": 1455524225.0, + "step": 2785 + }, + { + "epoch": 2.598043782021425, + "grad_norm": 0.19097546345147012, + "learning_rate": 7.4042112530203655e-06, + "loss": 0.1966, + "num_tokens": 1458115944.0, + "step": 2790 + }, + { + "epoch": 2.6027014438751745, + "grad_norm": 0.29883914842618964, + "learning_rate": 7.317915084570245e-06, + "loss": 0.1917, + "num_tokens": 1460737384.0, + "step": 2795 + }, + { + "epoch": 2.607359105728924, + "grad_norm": 0.19676602110755637, + "learning_rate": 7.231618916120124e-06, + "loss": 0.1933, + "num_tokens": 1463358824.0, + "step": 2800 + }, + { + "epoch": 2.6120167675826735, + "grad_norm": 0.1990332465486848, + "learning_rate": 7.1453227476700035e-06, + "loss": 0.1949, + "num_tokens": 1465980264.0, + "step": 2805 + }, + { + "epoch": 2.616674429436423, + "grad_norm": 0.23709753658921823, + "learning_rate": 7.059026579219883e-06, + "loss": 0.189, + "num_tokens": 1468590912.0, + "step": 2810 + }, + { + "epoch": 2.6213320912901725, + "grad_norm": 0.18992036833544143, + "learning_rate": 6.9727304107697625e-06, + "loss": 0.1774, + "num_tokens": 1471212352.0, + "step": 2815 + }, + { + "epoch": 2.625989753143922, + "grad_norm": 0.2068681005516739, + "learning_rate": 6.886434242319641e-06, + "loss": 0.1919, + "num_tokens": 1473833792.0, + "step": 2820 + }, + { + "epoch": 2.630647414997671, + "grad_norm": 0.20606667722426614, + "learning_rate": 6.800138073869521e-06, + "loss": 0.194, + "num_tokens": 1476455232.0, + "step": 2825 + }, + { + "epoch": 2.6353050768514206, + "grad_norm": 0.23982760911056664, + "learning_rate": 6.713841905419399e-06, + "loss": 0.1854, + "num_tokens": 1479076672.0, + "step": 2830 + }, + { + "epoch": 2.63996273870517, + "grad_norm": 0.21073555793675697, + "learning_rate": 6.627545736969279e-06, + "loss": 0.1949, + "num_tokens": 1481698112.0, + "step": 2835 + }, + { + "epoch": 2.6446204005589196, + "grad_norm": 0.19453233190357822, + "learning_rate": 6.541249568519157e-06, + "loss": 0.1981, + "num_tokens": 1484295121.0, + "step": 2840 + }, + { + "epoch": 2.6492780624126686, + "grad_norm": 0.28615224533623485, + "learning_rate": 6.454953400069037e-06, + "loss": 0.1843, + "num_tokens": 1486916561.0, + "step": 2845 + }, + { + "epoch": 2.653935724266418, + "grad_norm": 0.19196899236511794, + "learning_rate": 6.368657231618916e-06, + "loss": 0.1939, + "num_tokens": 1489515514.0, + "step": 2850 + }, + { + "epoch": 2.6585933861201676, + "grad_norm": 0.1974018590040202, + "learning_rate": 6.282361063168796e-06, + "loss": 0.1785, + "num_tokens": 1492136954.0, + "step": 2855 + }, + { + "epoch": 2.663251047973917, + "grad_norm": 0.19591164104142184, + "learning_rate": 6.196064894718675e-06, + "loss": 0.1913, + "num_tokens": 1494717079.0, + "step": 2860 + }, + { + "epoch": 2.6679087098276666, + "grad_norm": 0.19833670363700637, + "learning_rate": 6.109768726268554e-06, + "loss": 0.1892, + "num_tokens": 1497338519.0, + "step": 2865 + }, + { + "epoch": 2.672566371681416, + "grad_norm": 0.20511862285572172, + "learning_rate": 6.023472557818433e-06, + "loss": 0.1875, + "num_tokens": 1499959959.0, + "step": 2870 + }, + { + "epoch": 2.6772240335351656, + "grad_norm": 0.2007161163079071, + "learning_rate": 5.937176389368312e-06, + "loss": 0.192, + "num_tokens": 1502573694.0, + "step": 2875 + }, + { + "epoch": 2.6818816953889146, + "grad_norm": 0.21203335721803362, + "learning_rate": 5.850880220918191e-06, + "loss": 0.1995, + "num_tokens": 1505152753.0, + "step": 2880 + }, + { + "epoch": 2.686539357242664, + "grad_norm": 0.1972444939219368, + "learning_rate": 5.76458405246807e-06, + "loss": 0.197, + "num_tokens": 1507745161.0, + "step": 2885 + }, + { + "epoch": 2.6911970190964136, + "grad_norm": 0.18813496250520612, + "learning_rate": 5.67828788401795e-06, + "loss": 0.1897, + "num_tokens": 1510366601.0, + "step": 2890 + }, + { + "epoch": 2.695854680950163, + "grad_norm": 0.21683948817382884, + "learning_rate": 5.591991715567829e-06, + "loss": 0.1796, + "num_tokens": 1512988041.0, + "step": 2895 + }, + { + "epoch": 2.700512342803912, + "grad_norm": 0.20558268991109876, + "learning_rate": 5.5056955471177085e-06, + "loss": 0.1869, + "num_tokens": 1515609481.0, + "step": 2900 + }, + { + "epoch": 2.7051700046576617, + "grad_norm": 0.21816979196996153, + "learning_rate": 5.4193993786675876e-06, + "loss": 0.1803, + "num_tokens": 1518230921.0, + "step": 2905 + }, + { + "epoch": 2.709827666511411, + "grad_norm": 0.1974560996544184, + "learning_rate": 5.333103210217467e-06, + "loss": 0.1875, + "num_tokens": 1520838211.0, + "step": 2910 + }, + { + "epoch": 2.7144853283651607, + "grad_norm": 0.2142391911040735, + "learning_rate": 5.246807041767346e-06, + "loss": 0.1913, + "num_tokens": 1523450790.0, + "step": 2915 + }, + { + "epoch": 2.71914299021891, + "grad_norm": 0.19519798254614276, + "learning_rate": 5.160510873317225e-06, + "loss": 0.1898, + "num_tokens": 1526072230.0, + "step": 2920 + }, + { + "epoch": 2.7238006520726596, + "grad_norm": 0.19984436672708594, + "learning_rate": 5.074214704867105e-06, + "loss": 0.1927, + "num_tokens": 1528678779.0, + "step": 2925 + }, + { + "epoch": 2.728458313926409, + "grad_norm": 0.19080800820329621, + "learning_rate": 4.987918536416984e-06, + "loss": 0.1876, + "num_tokens": 1531300219.0, + "step": 2930 + }, + { + "epoch": 2.7331159757801586, + "grad_norm": 0.19259051393816032, + "learning_rate": 4.901622367966863e-06, + "loss": 0.1883, + "num_tokens": 1533921659.0, + "step": 2935 + }, + { + "epoch": 2.7377736376339077, + "grad_norm": 0.20045802931215237, + "learning_rate": 4.815326199516742e-06, + "loss": 0.1921, + "num_tokens": 1536543099.0, + "step": 2940 + }, + { + "epoch": 2.742431299487657, + "grad_norm": 0.20839258200951158, + "learning_rate": 4.729030031066621e-06, + "loss": 0.1995, + "num_tokens": 1539159628.0, + "step": 2945 + }, + { + "epoch": 2.7470889613414067, + "grad_norm": 0.19936373974638227, + "learning_rate": 4.6427338626165e-06, + "loss": 0.1946, + "num_tokens": 1541781068.0, + "step": 2950 + }, + { + "epoch": 2.751746623195156, + "grad_norm": 0.22149358888220214, + "learning_rate": 4.556437694166379e-06, + "loss": 0.2, + "num_tokens": 1544402508.0, + "step": 2955 + }, + { + "epoch": 2.7564042850489052, + "grad_norm": 0.1998628890141551, + "learning_rate": 4.470141525716258e-06, + "loss": 0.1955, + "num_tokens": 1547023948.0, + "step": 2960 + }, + { + "epoch": 2.7610619469026547, + "grad_norm": 0.19675965152173985, + "learning_rate": 4.383845357266138e-06, + "loss": 0.1953, + "num_tokens": 1549645388.0, + "step": 2965 + }, + { + "epoch": 2.765719608756404, + "grad_norm": 0.19054438082125744, + "learning_rate": 4.297549188816017e-06, + "loss": 0.1935, + "num_tokens": 1552266828.0, + "step": 2970 + }, + { + "epoch": 2.7703772706101537, + "grad_norm": 0.2100247956087177, + "learning_rate": 4.211253020365896e-06, + "loss": 0.1869, + "num_tokens": 1554875512.0, + "step": 2975 + }, + { + "epoch": 2.775034932463903, + "grad_norm": 0.22516581326706617, + "learning_rate": 4.124956851915775e-06, + "loss": 0.1829, + "num_tokens": 1557496952.0, + "step": 2980 + }, + { + "epoch": 2.7796925943176527, + "grad_norm": 0.22275259848789913, + "learning_rate": 4.0386606834656544e-06, + "loss": 0.1945, + "num_tokens": 1560118392.0, + "step": 2985 + }, + { + "epoch": 2.784350256171402, + "grad_norm": 0.2035646672810568, + "learning_rate": 3.9523645150155335e-06, + "loss": 0.1892, + "num_tokens": 1562730863.0, + "step": 2990 + }, + { + "epoch": 2.7890079180251512, + "grad_norm": 0.19893135307571172, + "learning_rate": 3.8660683465654126e-06, + "loss": 0.1917, + "num_tokens": 1565302118.0, + "step": 2995 + }, + { + "epoch": 2.7936655798789007, + "grad_norm": 0.1994424746191013, + "learning_rate": 3.779772178115292e-06, + "loss": 0.1907, + "num_tokens": 1567923558.0, + "step": 3000 + }, + { + "epoch": 2.7983232417326502, + "grad_norm": 0.1941087329302925, + "learning_rate": 3.693476009665171e-06, + "loss": 0.1897, + "num_tokens": 1570544998.0, + "step": 3005 + }, + { + "epoch": 2.8029809035863997, + "grad_norm": 0.19337805581200604, + "learning_rate": 3.6071798412150506e-06, + "loss": 0.185, + "num_tokens": 1573166438.0, + "step": 3010 + }, + { + "epoch": 2.807638565440149, + "grad_norm": 0.19861773096632682, + "learning_rate": 3.5208836727649297e-06, + "loss": 0.1807, + "num_tokens": 1575787878.0, + "step": 3015 + }, + { + "epoch": 2.8122962272938983, + "grad_norm": 0.20119520009288108, + "learning_rate": 3.434587504314809e-06, + "loss": 0.1941, + "num_tokens": 1578409318.0, + "step": 3020 + }, + { + "epoch": 2.8169538891476478, + "grad_norm": 0.1901334131646749, + "learning_rate": 3.348291335864688e-06, + "loss": 0.1881, + "num_tokens": 1581003943.0, + "step": 3025 + }, + { + "epoch": 2.8216115510013973, + "grad_norm": 0.2037752073186323, + "learning_rate": 3.2619951674145674e-06, + "loss": 0.1846, + "num_tokens": 1583601169.0, + "step": 3030 + }, + { + "epoch": 2.8262692128551468, + "grad_norm": 0.19225519448717485, + "learning_rate": 3.1756989989644464e-06, + "loss": 0.1858, + "num_tokens": 1586222609.0, + "step": 3035 + }, + { + "epoch": 2.8309268747088963, + "grad_norm": 0.19973043168505442, + "learning_rate": 3.089402830514325e-06, + "loss": 0.1946, + "num_tokens": 1588841357.0, + "step": 3040 + }, + { + "epoch": 2.8355845365626458, + "grad_norm": 0.18493025307293057, + "learning_rate": 3.0031066620642046e-06, + "loss": 0.1861, + "num_tokens": 1591462797.0, + "step": 3045 + }, + { + "epoch": 2.840242198416395, + "grad_norm": 0.2135184210397531, + "learning_rate": 2.9168104936140837e-06, + "loss": 0.1858, + "num_tokens": 1594081215.0, + "step": 3050 + }, + { + "epoch": 2.8448998602701443, + "grad_norm": 0.2040789058976081, + "learning_rate": 2.8305143251639627e-06, + "loss": 0.1899, + "num_tokens": 1596671205.0, + "step": 3055 + }, + { + "epoch": 2.849557522123894, + "grad_norm": 0.20810721325069814, + "learning_rate": 2.7442181567138422e-06, + "loss": 0.1998, + "num_tokens": 1599292645.0, + "step": 3060 + }, + { + "epoch": 2.8542151839776433, + "grad_norm": 0.19535763016764432, + "learning_rate": 2.6579219882637213e-06, + "loss": 0.1958, + "num_tokens": 1601907286.0, + "step": 3065 + }, + { + "epoch": 2.858872845831393, + "grad_norm": 0.19596015876020706, + "learning_rate": 2.5716258198136004e-06, + "loss": 0.1816, + "num_tokens": 1604525687.0, + "step": 3070 + }, + { + "epoch": 2.863530507685142, + "grad_norm": 0.20927607583564214, + "learning_rate": 2.4853296513634795e-06, + "loss": 0.1881, + "num_tokens": 1607147127.0, + "step": 3075 + }, + { + "epoch": 2.8681881695388913, + "grad_norm": 0.19480398108677532, + "learning_rate": 2.399033482913359e-06, + "loss": 0.1847, + "num_tokens": 1609768567.0, + "step": 3080 + }, + { + "epoch": 2.872845831392641, + "grad_norm": 0.18059331005524973, + "learning_rate": 2.312737314463238e-06, + "loss": 0.186, + "num_tokens": 1612390007.0, + "step": 3085 + }, + { + "epoch": 2.8775034932463903, + "grad_norm": 0.28380582647260677, + "learning_rate": 2.226441146013117e-06, + "loss": 0.1965, + "num_tokens": 1615000534.0, + "step": 3090 + }, + { + "epoch": 2.88216115510014, + "grad_norm": 0.18519192829599476, + "learning_rate": 2.140144977562996e-06, + "loss": 0.1731, + "num_tokens": 1617621974.0, + "step": 3095 + }, + { + "epoch": 2.8868188169538893, + "grad_norm": 0.18943911462800966, + "learning_rate": 2.0538488091128757e-06, + "loss": 0.1932, + "num_tokens": 1620243414.0, + "step": 3100 + }, + { + "epoch": 2.891476478807639, + "grad_norm": 0.20185510901537124, + "learning_rate": 1.9675526406627547e-06, + "loss": 0.1858, + "num_tokens": 1622864854.0, + "step": 3105 + }, + { + "epoch": 2.896134140661388, + "grad_norm": 0.1849272809577793, + "learning_rate": 1.8812564722126338e-06, + "loss": 0.1849, + "num_tokens": 1625486294.0, + "step": 3110 + }, + { + "epoch": 2.9007918025151374, + "grad_norm": 0.18974926125584826, + "learning_rate": 1.794960303762513e-06, + "loss": 0.1846, + "num_tokens": 1628107734.0, + "step": 3115 + }, + { + "epoch": 2.905449464368887, + "grad_norm": 0.19805184324376912, + "learning_rate": 1.7086641353123924e-06, + "loss": 0.1891, + "num_tokens": 1630709321.0, + "step": 3120 + }, + { + "epoch": 2.9101071262226363, + "grad_norm": 0.1923784438938438, + "learning_rate": 1.6223679668622715e-06, + "loss": 0.1917, + "num_tokens": 1633298077.0, + "step": 3125 + }, + { + "epoch": 2.9147647880763854, + "grad_norm": 0.18414721849685736, + "learning_rate": 1.5360717984121505e-06, + "loss": 0.1816, + "num_tokens": 1635910626.0, + "step": 3130 + }, + { + "epoch": 2.919422449930135, + "grad_norm": 0.19255074604870911, + "learning_rate": 1.4497756299620296e-06, + "loss": 0.1962, + "num_tokens": 1638532066.0, + "step": 3135 + }, + { + "epoch": 2.9240801117838844, + "grad_norm": 0.189954260071868, + "learning_rate": 1.363479461511909e-06, + "loss": 0.193, + "num_tokens": 1641153506.0, + "step": 3140 + }, + { + "epoch": 2.928737773637634, + "grad_norm": 0.20538589440314248, + "learning_rate": 1.277183293061788e-06, + "loss": 0.1965, + "num_tokens": 1643773175.0, + "step": 3145 + }, + { + "epoch": 2.9333954354913834, + "grad_norm": 0.19336904650694964, + "learning_rate": 1.1908871246116673e-06, + "loss": 0.186, + "num_tokens": 1646380875.0, + "step": 3150 + }, + { + "epoch": 2.938053097345133, + "grad_norm": 0.26726325995033184, + "learning_rate": 1.1045909561615463e-06, + "loss": 0.1883, + "num_tokens": 1648974711.0, + "step": 3155 + }, + { + "epoch": 2.9427107591988824, + "grad_norm": 0.19928547284660958, + "learning_rate": 1.0182947877114256e-06, + "loss": 0.1908, + "num_tokens": 1651596151.0, + "step": 3160 + }, + { + "epoch": 2.9473684210526314, + "grad_norm": 0.1999931278950155, + "learning_rate": 9.319986192613048e-07, + "loss": 0.1963, + "num_tokens": 1654217591.0, + "step": 3165 + }, + { + "epoch": 2.952026082906381, + "grad_norm": 0.18857341165059974, + "learning_rate": 8.45702450811184e-07, + "loss": 0.1836, + "num_tokens": 1656839031.0, + "step": 3170 + }, + { + "epoch": 2.9566837447601304, + "grad_norm": 0.21091493599043323, + "learning_rate": 7.594062823610632e-07, + "loss": 0.1924, + "num_tokens": 1659460471.0, + "step": 3175 + }, + { + "epoch": 2.96134140661388, + "grad_norm": 0.18915430253259063, + "learning_rate": 6.731101139109423e-07, + "loss": 0.1798, + "num_tokens": 1662081911.0, + "step": 3180 + }, + { + "epoch": 2.9659990684676294, + "grad_norm": 0.19423661318275964, + "learning_rate": 5.868139454608215e-07, + "loss": 0.1915, + "num_tokens": 1664702444.0, + "step": 3185 + }, + { + "epoch": 2.9706567303213784, + "grad_norm": 0.1987475010895318, + "learning_rate": 5.005177770107007e-07, + "loss": 0.1819, + "num_tokens": 1667317968.0, + "step": 3190 + }, + { + "epoch": 2.975314392175128, + "grad_norm": 0.21052305975759994, + "learning_rate": 4.142216085605799e-07, + "loss": 0.1842, + "num_tokens": 1669939408.0, + "step": 3195 + }, + { + "epoch": 2.9799720540288774, + "grad_norm": 0.18964427486372967, + "learning_rate": 3.279254401104591e-07, + "loss": 0.1879, + "num_tokens": 1672514586.0, + "step": 3200 + }, + { + "epoch": 2.984629715882627, + "grad_norm": 0.18975450375858877, + "learning_rate": 2.416292716603383e-07, + "loss": 0.1874, + "num_tokens": 1675136026.0, + "step": 3205 + }, + { + "epoch": 2.9892873777363764, + "grad_norm": 0.1909619964360884, + "learning_rate": 1.5533310321021747e-07, + "loss": 0.1757, + "num_tokens": 1677728036.0, + "step": 3210 + }, + { + "epoch": 2.993945039590126, + "grad_norm": 0.2036554562601111, + "learning_rate": 6.903693476009665e-08, + "loss": 0.1852, + "num_tokens": 1680349476.0, + "step": 3215 + }, + { + "epoch": 2.9976711690731253, + "num_tokens": 1682446628.0, + "step": 3219, + "total_flos": 2.754977641940386e+18, + "train_loss": 0.27190176190408577, + "train_runtime": 89221.753, + "train_samples_per_second": 0.577, + "train_steps_per_second": 0.036 + } + ], + "logging_steps": 5, + "max_steps": 3219, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 2.754977641940386e+18, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +}